This page intentionally left blank

JESUS AND GOSPEL

‘Gospel’ initially referred to oral proclamation concerning Jesus

Christ, but was later used to refer to four written accounts of the life
of Jesus. How did this happen? Here, distinguished scholar Graham
Stanton uses new evidence and fresh perspectives to tackle this con-
troversial question. He insists that in the early post-Easter period, the
Gospel of Jesus Christ was heard against the backdrop of a rival set of
‘gospels’ concerning the Roman emperors. In later chapters Stanton
examines the earliest criticisms of Jesus and of claims concerning his
resurrection. Finally, he discusses the early Christian addiction to the
codex (book) format as opposed to the ubiquitous roll, and under-
mines the view that early copies of the gospels were viewed as down-
market handbooks of an inward-looking sect. With half the material
previously unpublished and the rest carefully gathered from sources
difficult to access, this is a timely study with broad appeal.

Graham Stanton is Lady Margaret’s Professor of Divinity, University

of Cambridge, and a Fellow of Fitzwilliam College. His publications
include Gospel Truth? New Light on Jesus and the Gospels (1995), A
Gospel for a New People: Studies in Matthew (1992), The Gospels and
Jesus (1989: revised and expanded 2002) and Jesus of Nazareth in New
Testament Preaching (1974).
JESUS AND GOSPEL

GRAHAM N. STANTON
cambridge university press
Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo

Cambridge University Press

The Edinburgh Building, Cambridge cb2 2ru, UK
Published in the United States of America by Cambridge University Press, New York
www.cambridge.org
Information on this title: www.cambridge.org/9780521810326

© Graham N. Stanton 2004

This publication is in copyright. Subject to statutory exception and to the provision of

relevant collective licensing agreements, no reproduction of any part may take place
without the written permission of Cambridge University Press.

First published in print format 2004

isbn-13 978-0-511-21096-9 eBook (EBL)

isbn-10 0-511-21273-9 eBook (EBL)

isbn-13 978-0-521-81032-6 hardback

isbn-10 0-521-81032-9 hardback

isbn-13 978-0-521-00802-0 paperback

isbn-10 0-521-00802-6 paperback

Cambridge University Press has no responsibility for the persistence or accuracy of urls
for external or third-party internet websites referred to in this publication, and does not
guarantee that any content on such websites is, or will remain, accurate or appropriate.
Dedicated to Professor C. F. D. Moule
 Contents

Preface page ix
List of abbreviations xi

1 Introduction 1

part i jesus and gospel

2 Jesus and Gospel 9
3 The fourfold Gospel 63
4 Jesus traditions and gospels in Justin Martyr and Irenaeus 92
5 The law of Christ and the Gospel 110

part ii jesus
6 Jesus of Nazareth: a magician and a false prophet who deceived
God’s people? 127
7 Early objections to the resurrection of Jesus 148

part iii the gospels and papyri codices

8 Why were early Christians addicted to the codex? 165
9 What are the gospels? New evidence from papyri? 192

Bibliography 207
Index of passages cited 220
General index 233

vii
 Preface

The completion of a book is a time for stock-taking. Why have I written

on this topic, and not another? How have I managed to complete it, given
the ever-growing demands teaching and administrative duties make on the
time of an academic?
The topics explored in this book are at the very centre of the concerns of
anyone interested in earliest Christianity and, indeed, in Christian theology.
I have tried to approach them from fresh angles and, where possible, in the
light of new evidence. So I have spread my net more widely than is often
the case. The questions discussed have captured my interest for a variety of
reasons. In some cases I think that I have found new paths through well-
traversed territory. In others, I have become dissatisfied with the standard
answers.
Chapter 2, ‘Jesus and Gospel’, is a considerably extended version of the
Inaugural Lecture I gave as Lady Margaret’s Professor of Divinity in the
University of Cambridge on 27 April 2000. In my introductory remarks I
referred to the debt I owe to my two predecessors in the Cambridge Chair,
Professors C. F. D. Moule and Morna Hooker, who were both present.
An earlier version of Chapter 5, ‘The Law of Christ and the Gospel’,
was one of eighteen seminar papers given as part of the celebrations of
500th anniversary of the establishment of the Lady Margaret’s Professorship
in 1502. The seminar papers attempted to encourage dialogue between
Biblical scholars and theologians; they have now been published as Reading
Texts, Seeking Wisdom: Scripture and Theology, ed. David F. Ford and
Graham Stanton (London: SCM, 2003). Two lectures on the history of
the Professorship were given as the centrepiece of the celebrations; they are
included in Patrick Collinson, Richard Rex, and Graham Stanton, Lady
Margaret Beaufort and her Professors of Divinity at Cambridge (Cambridge:
Cambridge University Press, 2003).
Several of the other chapters are revised and extended versions of in-
vited lectures or seminar papers. The original settings have been diverse:
ix
x Preface
universities and colleges in New Zealand, Australia, Singapore, Canada,
the USA, Finland, France, and the Netherlands, as well as in the United
Kingdom. I am grateful for the initial invitations and the warm hospitality
received. Discussion following the lectures and seminars has often been en-
couraging. Sometimes new lines of inquiry have been suggested, and now
and again I have been forced to abandon false trails.
How have I managed to complete this book? I could not have done so
without the keen interest of my colleagues in Cambridge and earlier at
King’s College London. They have given me far more support and advice
than they are aware of. They have known when not to ask about progress,
and have often found ways of helping me to find time to press ahead. Only
a small number of my many doctoral students have been working with me
in the general field discussed in this book. But their enthusiasm for our
discipline, their lively questions, and their own promising scholarly work
have been a constant delight. My wife’s support has been unflagging; I hope
that I have not taken it for granted.
This book is dedicated to Professor C. F. D. Moule, the supervisor of
my doctoral research nearly forty years ago. His example of scholarly rigour
and his unswerving Christian commitment have meant more to me than
I can express. At ninety-five, he writes astute reviews for learned journals,
and by correspondence continues a ministry of encouragement to his many
friends, former colleagues and students.
 Abbreviations

ANRW Aufstieg und Niedergang der römischen Welt, ed.

H. Temporini and W. Haase (Berlin, 1972– ).
BDAG A Greek–English Lexicon of the New Testament and Other
Early Christian Literature, 3rd edn. revised and ed.
F. W. Danker (Chicago and London: University of Chicago
Press, 2000).
BJRL Bulletin of the John Rylands Library University of Manchester
CBQ Catholic Biblical Quarterly
CTQ Concordia Theological Quarterly
E. tr. English translation
EKK Evangelisch-katholischer Kommentar zum NT
ETL Ephemerides theologicae lovanienses
ExT Expository Times
FS Festschrift
ICC International Critical Commentary
JBL Journal of Biblical Literature
JSNT Journal for the Study of the New Testament
JTS Journal of Theological Studies
LCL Loeb Classical Library (Cambridge, Mass.: Harvard
University Press)
NovT Novum Testamentum
NTS New Testament Studies
REB Revised English Bible (1989)
RSR Religious Studies Review
SC Second Century
SJT Scottish Journal of Theology
ST Studia Theologica
TB Tyndale Bulletin

xi
xii List of abbreviations
TDNT Theological Dictionary of the New Testament (Grand Rapids:
Eerdmans, E. tr., 1964–76)
TS Theological Studies
Vig. Chr. Vigiliae Christianae
WUNT Wissenschaftliche Untersuchungen zum Neuen Testament
ZPE Zeitschrift für Papyrologie und Epigraphik
 chap t e r 1

Introduction

The main lines of inquiry pursued in this book are nearly all foreshadowed
in the lengthy, wide-ranging Chapter 2, ‘Jesus and Gospel’. Here I explore
the origin and the varied meanings of the ‘gospel’ word group all the way
from its use by Jesus to refer to his own proclamation to its use as the title
of a ‘book’ containing an account of the words and deeds of Jesus.
Although the term ‘gospel’ is as prominent in Christian vocabulary today
as it ever has been, there have been very few detailed studies in English of
the word group. It is difficult to account for the silence. Part of the answer
may lie in the onslaught James Barr launched in 1961 against the then
fashionable word studies.1 Only a fool would try to turn the clock back and
ignore Barr’s strictures. But I am not alone in thinking that it is now time
to reconsider some of the most important theological terms developed by
the earliest followers of Jesus. Of course, full attention must be given both
to the whole semantic field of which a given word group is part and to the
varied social and religious contexts in which it is used. I shall argue that,
when that is done, we find that, in the decade or so immediately after Easter,
followers of Jesus developed language patterns which differed sharply from
‘street’ usage in both the Jewish and the Graeco-Roman worlds. Some of
the terms which shaped early Christian theology were forged in ‘rivalry’
with contemporary language patterns. Scriptural themes and distinctive
Christian convictions played their part, but so too did dialogue with current
usage on the streets of east Mediterranean cities.
German scholars have been less coy about discussing the ‘gospel’ word
group. No doubt their interest has been encouraged by the prominence of
the terminology in the Lutheran tradition. Gerhard Friedrich’s important
article eÉagg”lion, first published in 1935 in the Theologisches Wörterbuch
zum Neuen Testament, drew on his teacher Julius Schniewind’s influential

1 James Barr, The Semantics of Biblical Language (London: SCM, 1961).

1
2 Introduction
study, Euangelion.2 Friedrich’s article is not immune from some of the criti-
cisms raised by James Barr, but it includes mountains of invaluable back-
ground material. I shall also refer to the major studies by Peter Stuhlmacher
(1968), Georg Strecker (1975), and Hubert Frankemölle (1994), sometimes
in disagreement, and in the later sections of my chapter I shall follow paths
none of these scholars has pursued.3
I shall suggest a quite specific setting in which Paul, his co-workers, and
his predecessors first began to use ‘gospel’ in ways at odds with current
usage. I shall insist that, although the imperial cult was not the source of
early Christian use of the word group, it was the background against which
distinctively Christian usage was forged and first heard. Christians claimed
that God’s once for all good news about Christ was to be differentiated from
Providence’s repeatable good news about the birth, accession, or return to
health of Roman emperors.
In the opening section of Chapter 2 I draw attention to the gap which is
opening up between the varied ways Christians use the ‘gospel’ word group
today and current secular usage. Sociolinguists have observed at first hand
the ways religious, political, ethnic, and other social groups develop their
own ‘insider’ terminology, often by adapting the vocabulary of ‘outsiders’.
So too in the first century. The first followers of Jesus developed their own
‘in-house’ language patterns, partly on the basis of Scripture, partly in
the light of their distinctive Christian convictions, but partly by way of
modifying contemporary ‘street’ language. I hope that this study of one
small part of the ‘social dialect’ of earliest Christianity will encourage simi-
lar studies, for this phenomenon seems to have escaped close attention
until now.
There is a further reason for focussing on the gospel word group. The
term ‘gospel’ is being used in some scholarly circles to provide legitimation
for particular views about the importance and authority of Q, the collection
of about 240 sayings of Jesus shared by Matthew and Luke. Q is now

2 G. Friedrich’s article eÉagg”lion was translated in G. Kittel, ed., Theological Dictionary of

the New Testament, Vol. ii (Grand Rapids: Eerdmans, E. tr. 1964), pp. 707–37. See also J.
Schniewind, Euangelion. Ursprung und erste Gestalt des Begriffs Evangelium, Vols. i–ii (Gütersloh,
1927/31).
3 P. Stuhlmacher, Das paulinische Evangelium (Göttingen: Vandenhoeck & Ruprecht, 1968); G. Strecker,
‘Das Evangelium Jesu Christi’, in G. Strecker, ed., Jesus Christus in Historie und Theologie, FS H.
Conzelmann (Tübingen: Mohr, 1975), pp. 503–48; H. Frankemölle, Evangelium. Begriff und Gattung,
2nd edn (Stuttgart: Katholisches Bibelwerk, 1994). Frankemölle’s book includes a helpful and very
full discussion of earlier literature.
 Introduction 3
referred to by some as a ‘gospel’,4 or as the ‘lost gospel’,5 in order to signal
that this hypothetical source is as important both for the historian and for
the theologian as the canonical gospels, Matthew, Mark, Luke, and John.
There is historical precedent for referring to Q as a ‘gospel’, for in the
second century some sets of diverse traditions concerning the life and teach-
ing of Jesus were referred to as ‘gospels’. But that precedent is beside the
point.6 In most references today to Q as a ‘gospel’, a different agenda is
at work. Modern portrayals of Jesus as a wisdom teacher on the basis of
an alleged original and largely historically reliable layer of Q traditions are
being offered as ‘good news’ to the post-modern world. What better way
of legitimating such views than by dubbing Q traditions ‘gospel’?7
So too with the exaggerated historical and theological claims made by
some on behalf of the Gospel of Thomas. In its present form it is a fourth-
century gnostic collection of sayings in Coptic attributed to Jesus. It is
now being referred to by some as ‘the fifth gospel’ in order to shore up
claims that its earlier layers provide access to a Jesus more congenial today
than the Jesus portrayed by New Testament writers as God’s good news for
humankind.8
So in spite of the second-century precedent for referring to diverse col-
lections of Jesus traditions as ‘gospels’, the assumption in some circles that
Q and Thomas are ‘gospel’ for humankind today is to be repudiated. The
primary reason for that is theological, not historical. Q and Thomas (and
several other apocryphal gospels) do contain valuable historical traditions,
but they do not proclaim the Gospel of Jesus Christ as witnessed to by Paul,
by Mark, and by other early Christians later deemed to belong to the circle

4 For a history of the use of ‘gospel’ for Q since 1988, see J. S. Kloppenborg Verbin, Excavating Q: The
History and Setting of the Sayings Gospel Q (Edinburgh: T. & T. Clark, 2000), p. 398 n. 63. See, for
example, R. A. Piper, ed., The Gospel Behind the Gospels: Current Studies on Q (Leiden: Brill, 1995);
J. M. Robinson, P. Hoffmann, and J. S. Kloppenborg, eds., The Sayings Gospel Q in Greek and in
English (Minneapolis: Fortress, 2002).
5 M. Borg et al., eds., The Lost Gospel Q: The Original Sayings of Jesus (Berkeley, Calif.: Ulysses, 1996).
6 F. Neirynck, a doyen Q specialist, still refuses to refer to Q as a ‘gospel’ on the grounds that it
is a hypothetical source; he prefers ‘the Sayings Source Q’. See ‘The Reconstruction of Q’, in A.
Lindemann, ed., The Sayings Source Q and the Historical Jesus (Leuven: Peeters, 2001), p. 57.
7 In effect this is conceded by Kloppenborg Verbin, Excavating Q, pp. 398–408. See also, for exam-
ple, R. W. Funk, Honest to Jesus: Jesus for a New Millennium (New York: HarperCollins, 1996);
R. W. Funk, ed., The Gospel of Jesus according to the Jesus Seminar (Sonoma, Calif.: Polebridge,
1999).
8 S. J. Patterson and J. M. Robinson, The Fifth Gospel Comes of Age (Harrisburg: Trinity Press, 1998);
Cf. N.T. Wright, ‘Five Gospels but No Gospel’, in B. Chilton and C. A. Evans, eds., Authenticating
the Activities of Jesus (Leiden: Brill, 1999), pp. 83–120.
4 Introduction
of apostles and their followers. When is a gospel not ‘Gospel’? When it is a
set of Jesus traditions out of kilter with the faith of the church. In essence,
this was Irenaeus’ answer at the end of the second century. I believe that it
still has theological validity today.
By now it will be apparent that consideration of the gospel word group
raises a whole set of historical and theological issues of perennial interest.
Towards the end of Chapter 2 (in 2.9) a particularly fascinating question
is discussed. When was ‘gospel’ first used to refer to a writing made up of
narratives about Jesus rather than to oral proclamation or its content? My
own answer is that the evangelist Matthew was the first to do so.
Once this new development in early Christian usage of the gospel word
group had taken place, further questions crowded in. How many ‘gospel
books’ did the church possess? Why did the second-century church even-
tually decide to fly in the teeth of critics who claimed that retention of four
inconsistent accounts of the life and teaching of Jesus undermined the credi-
bility of Christianity? What were the factors which led to Irenaeus’ classic
answer, ‘one Gospel in fourfold form’? Chapter 3 discusses the emergence
of the fourfold Gospel by drawing on many strands of evidence. The final
section of this chapter changes gear from historical to theological issues,
for acceptance of the fourfold Gospel carries with it several theological
implications.
Chapter 4 explores in detail one of the topics touched on in the previous
chapter. What status did Justin Martyr attach to the Jesus traditions and
the gospels he referred to in the middle of the second century? To what
extent does Irenaeus three decades or so later mark a break with Justin?
I emphasize more strongly than most scholars the importance of written
Jesus traditions for both Justin and Irenaeus.
In Chapter 5, the final chapter of Part I, I am still concerned with ‘Jesus
and Gospel’, but from a very different angle. I take as my starting point
Paul’s enigmatic phrase ‘the law of Christ’ (Gal. 6.2). I insist that for Paul
this ‘law’ is part of the Gospel he proclaimed, and not merely a slogan
used to refer to ethical teaching linked only loosely, if at all, to his major
theological concerns. I sketch the main ways this phrase and its cousins were
understood in early Christianity and in some parts of the later tradition.
Paul’s phrase needs considerable unpacking if it is to be of service to the
Christian Gospel today. When Paul’s understanding of ‘the law of Christ’
is complemented by the varied themes associated with this phrase and its
cousins up to the time of Justin Martyr, it can still enrich current theological
reflection. I remain a great admirer of the apostle Paul, but in this particular
case ‘earliest’ is not necessarily best. A canonical perspective helps, but some
 Introduction 5
of the most significant steps in interpretation of ‘the law of Christ’ were
taken in the second century.
In Chapters 6 and 7, the two chapters in Part II, I consider the earliest sets
of objections raised to the actions and teaching of Jesus, and to Christian
claims concerning his resurrection. The approach will seem to some to
be somewhat off-beat, and so it is. However, opponents of a political or
religious leader often see more clearly than followers what is at stake. So it
is entirely reasonable to search for polemical traditions. The quest is not
easy, for most of the anti-Jesus traditions have been preserved ‘against the
grain’ within early Christian writings.
Contemporary opponents of Jesus perceived him to be a disruptive
threat to social and religious order. His proclamation of God’s kingly rule
and its implications was rightly seen to be radical. For some, his teaching
and actions were so radical that they had to be undermined by an alter-
native explanation of their source. Jesus, it was claimed in his lifetime,
was a demon-possessed magician, and probably also a demon-possessed
false prophet. Readers who are au fait with the flood of recent literature
on the so-called historical Jesus will recognize that this is a conclusion
which runs against the tide. But I do not repent: I believe that it is well
founded.
There is an intriguing parallel with one of the key points made in
Chapter 2. From very early in the post-Easter period, proclamation of
the Gospel of Jesus Christ was heard against the backdrop of a rival set
of ‘gospels’ concerning the Roman emperors. The key question was this:
whose gospel? Providence’s provision of the emperor as saviour and bene-
factor, or God’s provision of Jesus Christ as redeemer and life-giver? Already
in the lifetime of Jesus there were rival answers on offer to the question:
who is this Jesus of Nazareth? For some he was in league with Beelzebul,
for others he was proclaiming in word and action God’s good news to the
poor as a messianic prophet. Both before and after Easter, followers of Jesus
rested their claims concerning him on their convictions concerning God,
and the relationship of Jesus to God.
The two chapters in Part III are, both, concerned with the earliest sur-
viving written traditions concerning Jesus Christ. Even though the earliest
papyri of the gospels are all quite fragmentary, they are of special interest,
for they are the earliest material evidence we have for Christianity.
In the past five years more very fragmentary papyri in the codex format
have become available. They confront us with the pressing questions which
are tackled in Chapters 8 and 9. Why are the earliest fragments of Christian
writings all in the unfashionable codex format? And do those early papyri
6 Introduction
tell us anything about the status and use of the writings in the Christian
communities which preserved them?
Chapter 8 asks why early Christians were addicted to the codex. I tackle
this question in some detail, and partly in the light of new evidence. I
differentiate three stages in early Christian use of the codex. My stage 3
concerns c. ad 300, the point at which Christian scribes’ addiction to the
codex may have first influenced non-Christian scribes. My stage 2 discusses
the variety of pragmatic factors which sustained early Christian addiction
to the codex. I then turn to stage 1, the initial precocious use of the codex
by scribes copying Christian writings.
My own insistence that in very earliest Christianity there was an almost
seemless transition from ‘notebook’ to ‘codex’ will seem blindingly obvious
to some, but in fact this explanation differs markedly from the ‘big bang’
theories on offer at present. If use of the codex was an extension of the use
of notebooks, then there are important corollaries: notebooks were used
by the very first followers of Jesus for excerpts from Scripture, for drafts
and copies of letters, and perhaps even for the transmission of some Jesus
traditions.
Chapter 9 claims that the recently published papyri of the gospels under-
mine the often-repeated view that, in contrast to Jewish copies of Scripture,
early copies of the gospels were the ‘workaday’, ‘utilitarian’, ‘downmarket’
handbooks of an inward-looking sect. The earliest surviving papyri of the
gospels confirm that, by the later decades of the second century, if not
earlier, the latter’s literary qualities and their authoritative status for the life
and faith of the church were widely recognized.

In this book I frequently try to build up a cumulative case on the basis of

as many strands of evidence as possible. Too much current New Testament
research is confined within ever smaller circles. Whenever the pot of familiar
questions is stirred repeatedly without the addition of new ingredients, the
resulting fare is both bland and predictable.
In nearly every chapter I have worked backwards from later, clearer
evidence and formulations to earlier, often partly hidden roots. Of course,
anachronism lurks at every corner, but disciplined use of this approach can
open up sorely needed fresh perspectives.
The origins of most books are complex. This one is no exception.
Chapters 2, 8, and 9 make up over half the book; only a handful of para-
graphs in these chapters have been published before. Chapters 3–7 are re-
vised and in some cases extended versions of earlier publications. Details of
the original publications are given at appropriate points in the notes.
 part i

Jesus and Gospel

 chap t e r 2

Jesus and Gospel

The subject of this chapter is the origin and early Christian use of the
noun ‘gospel’, the verb ‘to proclaim good news’ (or, ‘to gospel’), and a set
of near-synonyms.1 Given its importance in earliest Christianity and for
Christian theology more generally, discussion of this topic has not been as
extensive as one might have expected.2 On several key points opinion has
been keenly divided and no consensus has emerged. I shall revisit some of
the disputed issues and hope to advance discussion by offering several fresh
considerations. In particular, I shall focus on the function of the word group
in the religious and social setting of the earliest Christian communities.

2 .1 ‘gospel’ in current usage

In the sixteenth century the term ‘gospel’ featured frequently in the lan-
guage repertoire of Erasmus and the Reformers. Erasmus often referred to
‘the gospel philosophy’. In his ‘Prologue to the New Testament’ (1525) the
translator William Tyndale included an astute summary of ‘gospel’:
Euagelio (that we cal gospel) is a greke worde,
and signyfyth good, mery, glad and joyfull tydings,
that maketh a mannes hert glad,
and maketh him synge, daunce and leepe for ioye.3
In that tumultuous century the term ‘the gospel’ often functioned as a
shorthand way of referring to the Reformers and their distinctive views.
1 This chapter is a considerably extended version of my Inaugural Lecture as Lady Margaret’s Professor
of Divinity in the University of Cambridge, given on 27 April 2000.
2 There have been several major studies in German; details were given above, p. 2 nn 2–3. The word
group has attracted curiously little attention from English-speaking scholars, though a notable ex-
ception is the Australian ancient historian G. H. R. Horsley’s discussion, ‘The “Good News” of
a Wedding’, in New Documents Illustrating Early Christianity, Vol. iii (Macquarie University: The
Ancient History Documentary Research Centre, 1983), pp. 10–15. See also A. J. Spallek, ‘The Origin
and Meaning of EÉagg”lion in the Pauline Corpus’, CTQ 57 (1993) 177–90.
3 I owe this reference to R. I. Deibert, Mark (Louisville: Kentucky, 1999), p. 6.

9
10 Jesus and Gospel
For example, in 1547 John Hooper noted in a letter that, if the emperor
(Charles V) should be defeated in war, King Henry VIII would adopt ‘the
gospel of Christ’. ‘Should the gospel [i.e. the German Lutheran princes of
the Schmalkdic League] sustain disaster, then he will preserve his ungodly
masses.’4 In section 2.8 of this chapter we shall see that in the first century the
term ‘Gospel’ functioned similarly, as a shorthand term and as an identity
marker.
In recent decades ‘gospel’ has been commandeered with increasing fre-
quency by all colours and shades of Christians. Not long ago I discovered a
church in Canada which calls itself not simply ‘The Full Gospel Church’, a
tag I knew, but ‘The Four Square Gospel Church’. I have noticed that Pope
John Paul II likes the word ‘gospel’.5 In order to be ecumenically and the-
ologically correct today, ‘gospel’ has to be sprinkled liberally in all manner
of theological and ecclesiastical statements. Authors of popular Christian
books also like to include the term in their book titles.6
In current Christian use ‘gospel’ is a shorthand term whose content is
construed in different ways. Although the term sends out varying signals
according to context, there are usually some lines of continuity with the
early Christians’ insistence that ‘the Gospel’ (t¼ eÉagg”lion) is God’s good
news concerning the life, death, and resurrection of Jesus Christ.
In sharp contrast, however, the noun is used today in common parlance
very differently. In ‘street’ language it has one primary sense: ‘gospel truth’
is a statement on which one can rely absolutely. A recent article in a UK
national newspaper about new developments in lie detectors carried this
caption: ‘Do you tell porkies or gospel truth?’ Not long ago our builder
gave me a timetable for planned alterations to our home and said, ‘Graham,
don’t take this as gospel truth!’
There is a curious irony about current use of ‘gospel’ or ‘gospel truth’ to
refer to a statement on which one can rely completely. In ‘street’ language
today the phrase is a secularized version of Paul’s use of the phrase ‘the
truth of the gospel’ in Gal. 2.5 and 14. Current usage is miles away from

4 See Diarmaid MacCulloch, Tudor Church Militant: Edward VI and the Protestant Reformation
(London: Allen Lane, the Penguin Press, 1999), p. 58.
5 For example: Pope John Paul II, Fides et Ratio (1998). ‘The Gospel is not opposed to any culture . . .
Cultures are not only not diminished by this encounter; rather, they are prompted to open themselves
to the newness of the Gospel’s truth and to be stirred by this truth to develop in new ways.’
6 My colleague Dr Julius Lipner has drawn my attention to a fascinating and very different use of ‘gospel’
in a book title: The Gospel of Sri Ramakrishna, translated and edited by Swami Nikhilandanda (New
York: Ramakrishna-Vivekanda Center, 1942). Sri Ramakrishna is one of the best-known modern
Hindu holy men. Dr Lipner notes that here ‘gospel’ is clearly a loan-word from Christianity as it
impinged on Indian culture in nineteenth-century Bengal.
 Jesus and Gospel 11
Paul’s rich and profoundly theological understanding of the phrase.7 There
is now a considerable gap between Christian and secular use of ‘gospel’.
Secular use of ‘gospel’ is gradually becoming more common. If that were
to continue, in some countries the distinctive Christian use of the word
group would be overshadowed by secular use and thus become part of
the ‘in-house’ language of somewhat marginalized minority groups of
Christians. ‘Gospel’ would then be a ‘sociolect’, to use the term now favo-
ured by sociolinguists.8 I shall suggest in section 2.8 that the word group
functioned in precisely this way in the first century.

2 .2 plot ting the pat h

In this lengthy chapter my main points will be developed along the fol-
lowing lines. In the next section I shall claim that, although Jesus used the
verb ‘to proclaim God’s good news’ and was strongly influenced in his own
messianic self-understanding by Isa. 61.1-2, he did not use the noun ‘gospel’.
I shall then consider several possible explanations for the origin of the
word group in the early post-Easter period. The most striking feature of
earliest Christian usage is the way ‘the Gospel’ rapidly became a set phrase
whose content could simply be assumed by Paul and his co-workers with-
out the need for further explanation. I shall suggest that use of the noun
probably first emerged in Greek-speaking Christian circles as a radical
‘Christianizing’ of both the limited Biblical and the more extensive con-
temporary usage. Although we cannot be certain about the precise origin
of the distinctive ways Christians used the word group, it is clear that they
developed in rivalry with the prominent use in the propaganda and ideol-
ogy of the imperial cult of this word group and a clutch of associated themes.
The latter point is most important. The rivalry between ‘the one Gospel
of Jesus Christ’ and ‘the gospels’ of the Caesars encompasses far more than
the use of the ‘gospel’ word group.
In section 2.5 I shall refer to the ways in which ancient historians have
made considerable strides in the last two decades or so in advancing our
knowledge of the imperial cult in the first century. I shall then discuss some
of the more important literary and epigraphical evidence.

7 The Revd Barbara Moss has suggested to me that current secular use of ‘gospel truth’ may derive from
the custom of swearing on the Bible in a law court to tell ‘the truth, the whole truth, and nothing
but the truth’.
8 Sociolinguists now differentiate between an ‘idiolect’ and a ‘sociolect’. The former is an individual’s
idiosyncratic pattern of language, while the latter is pattern of language specific to a group – it may
include new coinage of vocabulary or specialized use of ‘normal’ terms. See section 2.8 below.
12 Jesus and Gospel
In section 2.6 I shall cautiously suggest that Paul’s initial proclamation
and his subsequent letter to the Galatian churches may have been heard
against the backdrop of the all-pervasive religious and social influence of
the imperial cult in the Roman colonies of Pisidian Antioch, Iconium, and
Lystra. I shall then refer more briefly in section 2.7 to the possibility that
this may also have been the case in Thessalonica and Philippi. In section 2.8
the function of the word group as a shorthand term and as part of an early
Christian ‘sociolect’ or social dialect will be considered.
One of the most surprising developments in early Christian use of the
noun ‘gospel’ took place towards the end of the first century, or early in
the second. In Paul’s day, and for at least a decade later, ‘gospel’ was used
by Christians in the singular to refer solely to oral proclamation. A century
later (c. ad 160) Justin Martyr referred to written accounts of the life and
teaching of Jesus as ‘gospels’. At some earlier point ‘oral gospel’ became
‘written gospel’, and ‘gospel’ became ‘gospels’. When was the noun ‘gospel’
first used to refer to a writing? I shall argue in section 2.9 that the evangelist
Matthew first took this momentous step – not the evangelist Mark, and
not Marcion. In Chapter 3 I shall discuss the emergence of the fourfold
Gospel in the second century, and in Chapter 4 the first use of ‘gospels’ by
Justin Martyr.
In the conclusions in section 2.10 I shall refer to an aspect of the
sharp question which has haunted New Testament scholarship for the last
200 years: how much continuity is there between the proclamation of good
news by the prophet from Nazareth and post-Easter proclamation of Jesus
as God’s good news? Is there a measure of continuity in the use of the
‘gospel’ word group and related terms before and after Easter?
There is a further preliminary point to mention before we go any further.
I shall focus primarily on one word group, though strictly speaking I should
discuss the whole semantic field of words and phrases used in early Christian
writings to refer to the heralding of God’s good news concerning Jesus
Christ: e.g. ‘the word’ (¾ l»gov; t¼ ç¦ma), ‘proclamation’ (t¼ kžrugma),
‘the message’ (¡ ˆkož, e.g. Gal. 3.5) and ¡ ˆggel©a (I John 1.5),9 witness (t¼
martÅrion), and ‘the faith’ (¡ p©stiv, Gal. 1.23). Of these terms, ¾ l»gov,
‘the word’, is the most significant for my present purposes. As we shall see,
it is used by Paul, Mark, Matthew, Luke, and the authors of Hebrews and
Revelation almost synonymously with t¼ eÉagg”lion, ‘the Gospel’.
9 R. E. Brown translates ¡ ˆggel©a as ‘the gospel’ and suggests that it may be the technical Johannine
equivalent of t¼ eÉagg”lion. He also claims that, when the Johannine believers spoke about the
content of what we call the Gospel of John, they may have referred to it as the angelia (¡ ˆggel©a).
The Epistles of John, Anchor Bible (Garden City: Doubleday, 1982), p. 193.
 Jesus and Gospel 13
2.3 jesus’ use of the ‘gospel’ word group
Discussion of Jesus’ use of the word group must start with its use in the
Old Testament, for the importance of Scripture for Jesus himself cannot
be exaggerated. There are only six examples of the Hebrew noun ‘Gospel’
(be sorah). In two cases (II Sam. 4.10; 18.22) the noun means ‘the reward
for good news’; in four passages it refers to the ‘good news’ of deliverance
from the enemy (II Sam. 18.20, 25, 27; II Kgs. 7.9). In all six passages
theological or religious overtones are conspicuous by their absence – and
this is a surprise to most Christians nurtured on the term ‘gospel’.
In five of the passages just listed, the Septuagint renders the Hebrew
noun be sorah as ¡ eÉaggel©a, a word not found in the NT. The noun t¼
eÉagg”lion, which is so important in early Christian writings, is found
only once in the LXX, and then in the plural in II Sam 4.10.10 Here
David states that he restrained and then killed the man who had told him
that Saul was dead and thought that he was thereby bearing good news
(eÉaggeliz»menov). ‘This was how I had to reward him for bringing good
ù
news’ (ހ ›dei me doÓnai eÉagg”lia). The plural (t‡ eÉagg”lia) is not
found in the NT at all. So, rather unexpectedly, neither the Hebrew text
nor the LXX is the direct source of the NT use of the noun t¼ eÉagg”lion.
With the verbal forms, however, matters are very different. They are
found in a number of OT passages with the general sense ‘to announce’, and
in some they are accompanied by a clear theological note. At Ps. 40.10 and
68.11 the good news proclaimed concerns an act of Yahweh’s. At Deutero-
Isaiah 40.9; 52.7; 60.6 and 61.1 and the related Ps. 96.2-3 (Ps. 95.2-3 LXX)
there is a strong eschatological and universal note: the victory and kingly
rule of Yahweh is proclaimed as good news by his messenger-prophet.11
These passages form the backdrop to numerous NT passages, and in
particular to several Jesus traditions. We need not doubt that Jesus saw his
words and deeds as fulfilment of the opening verses of Isaiah 61. ‘He has
sent me to announce good news to the poor, i.e. to gospel the poor . . . to
comfort all who mourn.’ Indeed, I believe that this passage was the most
important part of Scripture for Jesus’ own self-understanding: not Isaiah
53 with its references to the so-called suffering servant, but Isaiah 61.12

10 See further Stuhlmacher, Das paulinische Evangelium, pp. 155–6. He notes two further examples in
variants, 2 Sam. 18.27 and 2 Sam. 18.31 (LXX = 2 Kings), but accepts that the LXX contains no
examples of theological usage of eÉagg”lion.
11 The differences between the Hebrew and the LXX repay close attention, but they do not affect the
general point being made here.
12 See C. M. Tuckett, ‘Scripture and Q’, in C. M. Tuckett, ed., The Scriptures in the Gospels (Leuven:
University Press, 1997), pp. 20–6.
14 Jesus and Gospel
The evangelist Luke certainly took this view. He opens his account of the
ministry of Jesus with that dramatic scene set in the synagogue in Nazareth.
Jesus stands up and reads the lesson, and is handed the scroll of the prophet
Isaiah. He opens the scroll and reads,

The spirit of the Lord is upon me

because he has anointed me to announce good news to the poor (eÉaggel©zesqai
ptwco±v),
to proclaim release for prisoners
and recovery of sight for the blind;
to let the broken victims go free,
to proclaim the year of the Lord’s favour.

Jesus rolled up the scroll, gave it back to the attendant, and sat down. Luke
adds, ‘all eyes in the synagogue were fixed on him’. Then Jesus addresses
those present: ‘Today, this scripture has been fulfilled in your hearing’ (Luke
4.16-21).
Now, in its present form this passage has undoubtedly been shaped by
Luke as a dramatic opening to his account of the ministry of Jesus, a
scene which is programmatic for his two volumes: many of Luke’s dis-
tinctive themes are foreshadowed in these verses. Nonetheless, the core
of this passage goes back to Jesus. I shall mention only two reasons for
taking this view. First, the two other passages in which Jesus refers to
Isaiah 61 (to which we shall turn in a moment) have even stronger claims
to historicity; the core of this passage coheres with them. Secondly, not
even Luke makes Christological capital out of this passage. It is often
overlooked that only in the scenes which follow in chapter 4 is Jesus
said (and then only by demons) to be the Holy One of God (4.34), the
Son of God, the Messiah (4.41). But in the Nazareth synagogue scene
Luke’s Jesus makes no more than an indirect claim that he himself is
the anointed prophet sent by God to announce good news to the poor.
The reticence of Jesus to claim that he himself is the content of the good
news (and not merely its proclaimer) is all of a piece with the evidence
elsewhere: this passage has not been deeply impregnated with post-Easter
Christology.
Isaiah 61 also plays an important role in the wording and themes of the
opening Beatitudes, both in Matthew and in Luke. ‘Blessed are you who
are poor, for yours is the kingdom of God’ might almost be paraphrased
as, ‘God is announcing good tidings of salvation to the poor’, for ‘blessed’
(mak†riov) echoes LXX usage, where it expresses the happiness which is the
 Jesus and Gospel 15
result of God-given salvation.13 The authenticity of the opening Beatitudes
and their close link with Isaiah 61 are generally agreed upon.14
Matthew opens the first of his five carefully constructed presentations of
the teaching of Jesus with the Beatitudes. In fact, I think it is very probable
that the evangelist Matthew extended the echoes of Isaiah 61 already present
in the tradition which came to him. So in Matthew’s Gospel, as well as
in Luke’s programmatic scene in the synagogue in Nazareth, Isaiah 61 is
prominent in the very first words spoken by Jesus. Quite independently,
and I think quite correctly, both evangelists discerned the importance of
this passage for Jesus himself.
I turn now to an important Q passage which I shall discuss in more detail:
the reply of Jesus to John the Baptist’s inquiry. The wording of Matt. 11.2-6
and the parallel passage in Luke 7.19, 22-3 are almost identical, so that the
underlying Q tradition can be set out without difficulty.
When John heard (in prison), he sent word by his disciples saying, ‘Are you the
one who is to come, or are we to expect someone else?’
And Jesus answered them, ‘Go and tell John what you have seen and heard:
The blind recover their sight, the lame walk,
lepers are cleansed, the deaf hear,
the dead are raised up, and the poor have the good news brought to them.
And blessed is anyone who takes no offence at me.’
Note how the list of the actions of Jesus comes to a climax with ‘the dead
are raised to life, the poor are brought good news’. With the exception
of ‘lepers are cleansed’, the items in the list are all allusions to phrases in
Isa. 29.18; 35.5-6, and 61.1-2. If we were writing out that list, we might be
inclined to place ‘the dead are raised to life’ as the dramatic conclusion.
And that is precisely the alteration to the order of the clauses made by
a few scribes.15 But the list reaches its climax with the clear allusion to
Isa. 61.1, ‘the poor are brought good news’, ‘the poor are gospelled’. Jesus
is claiming that both his actions and his proclamation of God’s good news
are fulfilment of Scriptural promises.

13 U. Luz, Das Evangelium nach Matthäus, EKK i/1, 5th edn (Düsseldorf and Zurich: Benziger, 2002),
pp. 276–7, discusses the problems which face the translator of mak†riov and concludes: ‘Eine ideale
Übersetzung gibt es im Deutschen nicht.’ The same is true in English.
14 The precise relationship of the opening beatitudes to Isa. 61.1-2, 7 is disputed. For a summary of
recent scholarship see Luz, Matthäus, pp. 271–2. For detailed discussion see F. Neirynck, ‘Q6, 20b–21;
7, 22 and Isaiah 61’, in C. M. Tuckett, ed., The Scriptures in the Gospels (Leuven: University Press,
1997), pp. 27–64.
15 Y, family 13 and a few other minuscules, the Curetonian Syriac.
16 Jesus and Gospel
One of the fragments of the so-called Messianic Apocalypse discovered
in Cave 4 at Qumran and known as 4Q521 provides a significant parallel
and sheds fresh light on the interpretation of this Q passage.
1 [for the heav]ens and the earth will listen to his anointed one, 2 [and all] that
is in them will not turn away from the precepts of the holy ones. 3 Strengthen
yourselves, you who are seeking the Lord, in his service! Blank 4 Will you not in
this encounter the Lord, all those who hope in their heart? 5 For the Lord will
consider the pious, and call the righteous by name, 6 and his spirit will hover upon
the poor, and he will renew the faithful with his strength. 7 For he will honour the
pious upon the throne of eternal kingdom, 8 freeing prisoners, giving sight to the
blind, straightening out the twis[ted]. 9 And for[e]ver shall I cling to [those who]
hope, and in his mercy [ . . . ] 10 and the fru[it of . . . ] . . . not be delayed. 11 And the
Lord will perform marvellous acts such as have not existed, just as he sa[id] 12 [for]
he will heal the badly wounded and will make the dead live, he will proclaim
good news to the poor 13 and [ . . . ] . . . [ . . . ] he will lead the [ . . . ] and enrich
the hungry. 14 [ . . . ] and all [ . . . ] (Frag. 2, col. II)
This is part of the largest of seventeen fragments from the writing first
published in 1992.16 Once again phrases from Isaiah are woven together. In
line 12 we find an astonishing parallel with the reply of Jesus to John. ‘He
will heal the wounded, give life to the dead and preach good news to the
poor.’ The order is identical: in both passages proclamation of good news
to the poor forms the climax of the list of actions to be carried out by God.
In both passages allusion to the fulfilment of Isa. 61.1 is unmistakable.
This fragment of 4Q521 opens with an almost certain reference to the
Messiah, ‘his anointed one’. In the lines which follow it is God who cares
for the various needy groups, and raises the dead. God does not usually
‘preach good news’; this is the task of his herald, messenger, or prophet.17
The herald or messenger referred to is the Messiah. So Isa. 61.1 is interpreted
messianically in this fragment.
There is further support in another Qumran fragment for this inter-
pretation. In lines 15 and 16 of 11Q13 (known earlier as 11Q Melchizedek)
Isa. 52.7 is quoted in full. The ‘messenger who announces peace, the mes-
senger of good who announces salvation’ is ‘the one anointed by the spirit’
16 The translation is taken from The Dead Sea Scrolls: Study Edition, edited and translated by Florentino
Garcı́a Martı́nez and Eibert J. C. Tigchelaar, Vol. ii (Brill: Leiden, 1998), p. 1045. The Hebrew text
is printed on the facing page, and a bibliography is included. I have supplied the bold type.
17 John J. Collins, The Scepter and the Star (New York: Doubleday, 1995), pp. 116–23. For discussion of
more recent literature and support for the view taken here, see J. J. Collins, ‘Jesus, Messianism and
the Dead Sea Scrolls’, in J. H. Charlesworth, H. Lichtenberger, and G. S. Oegema, eds., Qumran –
Messianism: Studies on the Messianic Expectations in the Dead Sea Scrolls (Tübingen: Mohr–Siebeck,
1998), pp. 100–19, esp. 112–16; C. A. Evans, ‘Jesus and the Dead Sea Scrolls’, in P. W. Flint and
J. C. Vanderkam, eds., The Dead Sea Scrolls after Fifty Years (Leiden: Brill, 1999), pp. 585–8.
 Jesus and Gospel 17
about whom it is written (Isa. 61.1-2) that he will proclaim ‘comfort to the
afflicted’. Although this passage is fragmentary and difficult to interpret in
detail, the herald of good tidings of Isa. 52.7 is closely linked with Isa. 61.1
and is identified as ‘the anointed one’, the Messiah.18
So we now have clear evidence that, before the time of Jesus, Isa. 61.1,
with its reference to the anointed prophet being sent to preach good news
to the poor, was understood to refer to a messianic prophet. It is highly
likely that, when Jesus referred to his own actions and words in terms of
this passage (and the related passages in Deutero-Isaiah), he was making an
indirect messianic claim. He was not merely a prophet proclaiming God’s
good news; he was himself part of the good news.
But what about the historicity of John’s question to Jesus, and the reply?
Two points strongly suggest that these verses are not simply a post-Easter
development. John asks, ‘Are you the one who is to come, or are we to expect
another?’ Jesus does not reply directly to this question. His refusal to make
overt claims about himself coheres with many other Jesus traditions and is
out of kilter with post-Easter tendencies.
Jesus leaves John’s disciples, and John himself, to work out the answer
to their question. John has heard about the actions and words of Jesus, and
asks about their significance. Jesus’ probing, teasing method of encouraging
his questioners to think through matters for themselves is all of a piece with
the parables. In the case of the parables there is general agreement that this
indirect method of communication is undoubtedly authentic. So too with
this passage.19
Note how it ends. ‘Blessed are those who take no offence at me.’ That
saying clearly implies that there were those who did take offence at the
actions and words of Jesus. We know from both Christian and Jewish
sources that Jesus was seen in his own lifetime to be a false prophet who
led Israel astray, a magician whose healings and exorcisms were the result of
collaboration with the prince of demons. So this passage raises the question
of the relationship of Jesus to God. Was Jesus a messianic prophet fulfilling
Isaiah 61 and proclaiming God’s good news to the poor? Or was he a false
prophet leading Israel astray? Jesus’ proclamation of God’s good news, his

18 For text and translation, with recent bibliography, see The Dead Sea Scrolls, Vol. ii, pp. 1206–9. For
earlier discussion and bibliography see G. N. Stanton, ‘On the Christology of Q’, in B. Lindars and
S. S. Smalley, eds., Christ and Spirit in the New Testament (Cambridge: Cambridge University Press,
1973), pp. 27–42.
19 See especially J. Ian H. McDonald, ‘Questioning and Discernment in Gospel Discourse: Commu-
nicative Strategy in Matthew 11.2-9’, in B. Chilton and C. A. Evans, eds., Authenticating the Words
of Jesus (Leiden: Brill, 1999), pp. 333–62.
18 Jesus and Gospel
gospelling, if you like, was in competition and dialogue with an alternative
story.20
Isaiah 61 is deeply embedded in the three passages I have referred to
briefly. Jesus’ proclamation of good news, of evangel, is in accordance with
Scripture and is its fulfilment. If we have ears to hear and eyes to see, then
it is possible to discern that Jesus himself is part of the proclamation.
But did Jesus use an Aramaic equivalent of the noun ‘gospel’ (be sorah)?
Here we face a puzzle. I have insisted that Jesus used the verb ‘to proclaim
good news’, but that verb is not used by Mark at all. Mark uses the noun t¼
eÉagg”lion in the absolute, five times on the lips of Jesus (1.15; 8.35; 10.29;
13.10; 14.9), but never the verb. The mystery deepens when we note that the
noun t¼ eÉagg”lion is not found either in Q traditions or in Luke’s or in
John’s Gospels. The bafflement continues when we discover that Matthew
omits three of Mark’s uses of t¼ eÉagg”lion on the lips of Jesus (Mark
1.15; 8.35; 10.29) and expands the other two (cf. Matt. 24.14 and 26.13 and
Mark 13.10 and 14.9). In other words, Matthew’s redactional hand has so
clearly reshaped radically Mark’s use of the noun that we cannot look to
this gospel for evidence of Jesus’ own usage.21 There is no other evidence
in the gospels directly relevant to our question.
So we must focus on the five examples of t¼ eÉagg”lion on the lips
of Jesus in Mark. The evangelist uses the noun in 1.1 and 1.14 as part of
the comments he makes as narrator on the significance of the story he is
unfolding. In both cases the noun is qualified: ‘the gospel of Jesus Christ’
(1.1) and ‘the gospel of God’ (1.14). The other five times in Mark are all in
the absolute, ‘the gospel’, without any qualifying phrase.
Do these five verses reflect Jesus’ own use of the noun, or post-Easter
terminology? The phraseology of Mark 8.35 and 10.29 is similar, ‘for my sake
and for the sake of the gospel’. Since ‘for the sake of the gospel’ (™neken toÓ
eÉaggel©ou) is not found in the parallel passages in Matthew and Luke, it
has often been suggested that in neither case was this phrase included in the
‘first’ edition of Mark used by the later evangelists.22 Whether or not that
was so may be left as an open question, but the phrase is an explanation or
interpretation of the preceding phrase, ‘for my sake’ (™neken –moÓ), using
post-Easter vocabulary. Jesus refers to himself as the content of ‘the gospel’ –
but elsewhere he is very reluctant to refer to himself as the content of his own
proclamation. With reference to these two passages, Willi Marxsen sums

20 See Chapter 6 below. 21 For fuller discussion see section 2.9 below.
22 So, for example, G. Friedrich in his influential article eÉagg”lion in TDNT ii, p. 727. W. Marxsen,
Mark the Evangelist (E. tr. Nashville: Abingdon, 1969), p. 124 considers this ‘highly improbable’.
 Jesus and Gospel 19
up the evangelist’s point: ‘Whoever suffers today [i.e. in Mark’s day] for the
gospel’s sake or abandons this world’s goods for the gospel’s sake does so
for the sake of the Lord.’23 ‘For the sake of Jesus Christ’ and ‘for the sake of
the gospel’ are all but synonymous expressions. Peter Stuhlmacher suggests,
surely correctly, that these two verses ‘point to the self-understanding and
sense of mission of early Christian missionaries to the Gentiles other than
Paul’.24
In Mark 13.9 Jesus warns his followers that they will be summoned to
appear before governors and rulers ‘on my account’ (13.9, ™neken –moÓ).
Verse 10 explains that this will happen as ‘the gospel is proclaimed to all
nations’, so a post-Easter setting is envisaged. With minimal alteration to
the sense, in verse 10 ‘the gospel’ could be replaced by ‘Jesus Christ’, and
in verse 9 ‘on my account’ could be replaced by ‘on account of the gospel’.
Hence 13.10 does not refer to Jesus’ own proclamation of the Gospel, but
to post-Easter proclamation of him as God’s good news.
In Mark 14.9 post-Easter proclamation of the Gospel throughout the
world is again in view: wherever the Gospel is proclaimed ‘in the whole
world’, the woman’s spontaneous act of devotion to Jesus in the house
of Simon the leper at Bethany will be told in remembrance of her. So in
none of these four passages (8.35; 10.29; 13.10; 14.9) is Jesus’ own pre-Easter
proclamation of good news clearly in view; in all four verses the evangelist
uses post-Easter phraseology.
The only further use of the noun ‘gospel’ in Mark on the lips of Jesus
raises a set of problems. Jesus proclaims, ‘Repent, and believe the gospel’
(1.15). In the preceding verse the evangelist as narrator states that Jesus had
come into Galilee proclaiming ‘the gospel of God ’ (t¼ eÉagg”lion toÓ
qeoÓ). But in verse 15 on the lips of Jesus (as in the other four verses just
discussed), the absolute term is used: ‘the gospel ’, t¼ eÉagg”lion; there are
no explanatory additional phrases. Unlike the other passages in which t¼
eÉagg”lion occurs, post-Easter proclamation to Gentiles is not envisaged.
In this summary of the proclamation of Jesus the ripples of distinctively
Christian (and especially Pauline)25 post-Easter use of the noun in the
absolute (i.e. ‘the gospel’) can be seen.
That judgement has been contested. Some scholars draw attention to
the unusual phrase, ‘believe in the gospel’, pisteÅete –n tä€ eÉaggel©w„,

23 Marxsen, Mark the Evangelist, p. 128.

24 P. Stuhlmacher, ‘The Theme: The Gospel and the Gospels’, in P. Stuhlmacher, ed., The Gospel and
the Gospels (Grand Rapids: Eerdmans, 1991), pp. 22–3, and esp. n. 81.
25 For careful discussion of Paul’s influence on Mark, see J. Marcus, ‘Mark – Interpreter of Paul’, NTS
46 (2000) 473–87.
20 Jesus and Gospel
and suggest a Semitic background. They then allow that it may go back
to Jesus.26 However, although this phrase is unusual in Greek, it is not
impossible.27 If it is taken as an authentic phrase used by Jesus, then it stands
in splendid isolation, for there is no parallel use of the noun elsewhere in the
Jesus traditions. On close analysis, the only part of Mark 1.15 with strong
claims to authenticity is ‘the kingdom of God has drawn near: repent’.28
If, as I have argued, the evangelist has used post-Easter phraseology, t¼
eÉagg”lion, in the summary of the proclamation of Jesus in 1.15 and in the
other four verses where the phrase is found on the lips of Jesus (8.35; 10.29;
13.10; 14.9); then there is a significant corollary. For the evangelist Mark,
‘the gospel preached by the church is identical with the gospel preached by
Jesus’.29
I have emphasized that Jesus understood his own proclamation and role
as a fulfilment of Isa. 61.1-2, with its reference to the anointed one who
proclaims God’s good news. But, since the noun ‘gospel’ is not used either
in the Hebrew or the LXX in Isa. 61, the origin of t¼ eÉagg”lion must be
sought elsewhere. Jesus himself used the verb ‘to proclaim good news’, but
not the noun. The five occurrences on the lips of Jesus in Mark are best
explained as post-Easter usage.

2 .4 the earliest christ ian usage

I turn now to the post-Easter period, and first of all to Paul. The noun
‘gospel’ is used sixty times in the Pauline letters, forty-eight times in the
undisputed letters. In just over half of those passages, t¼ eÉagg”lion is
used absolutely, i.e. without any additional explanatory phrase such as ‘of
God’ or ‘of Christ’. In the Pauline corpus and in Christian writings up to
Justin Martyr in the middle of the second century, the noun is always used
in the singular. The prominence of the noun in early Christian writings is
astonishing, especially given the fact that the noun is used only once in the
LXX at II Sam. 4.10 – and then in the plural, of the supposed ‘good news’
which turned out to be ‘bad news’. In non-Christian writings, inscriptions

26 For example, P. Stuhlmacher (‘The Theme: The Gospel and the Gospels’, pp. 20–1) suggests that
Mark 1.15, ‘with its very striking semitism, pisteÅete –n tå €eÉaggel©w„, believe in the gospel, is best
explained (as Schlatter had already observed) as tradition’. J. Marcus, Mark 1-8, Anchor Bible 27
(New York: Doubleday, 2000), p. 174, suggests that Mark 1.15 is a baptismal formula, though its gist
may go back to the historical Jesus.
27 See further C. F. D. Moule, An Idiom-Book of New Testament Greek (Cambridge: Cambridge Uni-
versity Press, 1960), pp. 80–1, 205.
28 So too J. P. Meier, A Marginal Jew, Vol. ii (New York: Doubleday, 1994), pp. 431 and 485.
29 Morna D. Hooker, The Gospel according to St Mark (London: A. & C. Black, 1991), p. 34.
 Jesus and Gospel 21
and papyri of this period, the noun is used reasonably frequently, but with
only a few possible exceptions to be discussed below, only in the plural.
The corresponding verb is less prominent in the Pauline corpus than the
noun; it is used in sixteen passages in the undisputed letters, and twenty-
one times in all. In six of the sixteen passages noun and verb are juxtaposed
(Rom. 10.15-16; I Cor. 9.18; 15.1; II Cor. 11.7; Gal. 1.6-8; 1.11), and in several
more the context makes it clear that proclamation of God’s good news
concerning Christ is in mind. In other words, even where the verb is used,
the sense conveyed by the noun is usually not far away.
Statistics are often misleading, and word statistics are no exception. But
in the case of the ‘gospel’ word group, they cry out for explanation. What
was the origin of the distinctive Christian usage of this word group? And
why was it used so frequently in early Christian circles? There are several
possible explanations.
We have already noted that early post-Easter usage of the noun was not
influenced by Jesus himself, for he does not appear to have used an Aramaic
equivalent. It is possible that some of the followers of Jesus drew on their
knowledge of extra-Biblical Jewish traditions in Hebrew or Aramaic in
developing the earliest Christian use of the noun. From time to time a
hypothesis along these lines has been proposed, but has won no more than
minimal support.30 The main problem concerns the dating of the phrases
in Aramaic targums or rabbinic traditions which are claimed as the back-
ground of early Christian use of the noun in Greek.31
Another possibility is that the distinctively Christian use of the noun was
‘coined’ on the back of post-Easter use of the verb, with its Biblical roots in
the LXX passages noted above. Given that the verb is used in a handful of
LXX passages to refer to proclamation of good news concerning Yahweh’s
action on behalf of his people, at first sight this seems to be a plausible
explanation. However, it is not entirely convincing.
The LXX passages which use the verb in a rich theological sense and
which might most readily be posited as the fons et origo of early Christian
usage of the word group are as difficult to find in Paul’s letters as a needle
in a haystack. The only partial exception is Paul’s reference to Isa. 52.7 in
Rom. 10.15: ‘As scripture says, “How welcome are the feet of the messengers
of good news!”’ (tän eÉaggelizom”nwn ˆgaq†).32 But the apostle is not

30 For full discussion see Frankemölle, Evangelium (above, Chapter 1, n. 3), pp. 76–86.
31 P. Stuhlmacher suggests cautiously that the Targum on Isa. 53.1 may be relevant, but he does not
attempt to date the Targum. The Gospel and the Gospels, p. 20 n. 74 and pp. 22–3 n. 22.
32 The extent of Paul’s adaptation of either the Masoretic Hebrew text or the LXX has been much
discussed. There now seems to be agreement that Paul is adapting a ‘non-standard’ Greek text. See
22 Jesus and Gospel
referring to part of Isa. 52.7 in order to underline its fulfilment in the
good news concerning the coming of Jesus Christ. Isaiah’s announcement
concerning the good news of Yahweh’s deliverance of his people and his
kingship is adapted by Paul in order to underline the irony that not all
have responded to the Gospel (Rom. 10.16). Hence Isa. 52.7 is not a strong
candidate in our quest for the origin of Christian usage of the ‘gospel’ word
group.
Isa. 61.1-2 is an even less likely candidate, in spite of the fact that, as
noted above, it was particularly important for Jesus’ own messianic self-
understanding. For in Paul’s extant writings this passage is not referred to
at all. So in our quest for the origin of Christian use of the word group, we
must look elsewhere and consider extra-Biblical usage.

2.4.1 When and where?

In order to pursue our quest for the origin of early Christian use of the
word group, we must seek as much precision as possible concerning the
date and geographical location of that development.
In his letters Paul regularly assumes that their recipients are thoroughly
familiar with the ‘gospel’ terminology he and his co-workers used. In several
passages he explicitly notes that he had used it in his initial preaching. At
I Thess. 1.5 Paul refers to ‘the gospel’ which he and his co-workers had
brought to the city of Thessalonica on his initial visit to that city. There
are similar references at Gal. 1.8-9,11 and 4.13 to Paul’s initial proclamation
in the Galatian churches as ‘the gospel’. So, long before Paul wrote to
Christians in Thessalonica and to the Galatian churches, the distinctive
Christian use of ‘the gospel’ was well established. But how much earlier?
In our quest for origins we must give priority to Paul’s own statements,
and also take seriously (but not uncritically) information provided by Luke
in Acts.33 Paul states that God’s disclosure of his Son to him, i.e. ‘the gospel’,
and God’s call to proclaim good news (¯na eÉaggel©zwmai) concerning him
among the Gentiles took place in, or near, Damascus (Gal. 1.14-17; see also
II Cor. 11.32). While Paul is claiming that the content of the proclamation

especially D.-A. Koch, Die Schrift als Zeuge des Evangeliums (Tübingen: Mohr, 1986), pp. 66–9, 81–
2, 113–14, 122; C.D. Stanley, Paul and the Language of Scripture (Cambridge: Cambridge University
Press, 1992), pp. 134–41; Shiu-Lun Shum, Paul’s Use of Isaiah in Romans (Tübingen: Mohr–Siebeck,
2002).
33 See especially Rainer Riesner’s critical appraisal of attempts to reconstruct the chronology of earliest
Christianity, Paul’s Early Period: Chronology, Mission Strategy, Theology (Grand Rapids: Eerdmans,
1998), pp. 3–32 (first published as Die Frühzeit des Apostels Paulus (Tübingen: Mohr, 1994)).
 Jesus and Gospel 23
he was called to deliver was disclosed to him on that occasion, that does
not necessarily apply to the noun, t¼ eÉagg”lion.
Three years later, following a visit to Arabia, Paul went up to Jerusalem
and saw Cephas (Peter) before travelling into the regions of Syria and Cilicia
for an extended visit, where he ‘announced good news concerning the
faith’ (eÉaggel©zetai tŸn p©stin, Gal. 1.21-4). Here ‘the faith’ is almost
synonymous with ‘the gospel’. That visit to Syria would naturally have
included Antioch, and there, surely, Paul preached to Gentiles in Greek,
as well as to Jews. Since there is general scholarly agreement that Paul’s
conversion or call took place in c. ad 33,34 Paul’s first visit to Jerusalem took
place in c. 36/7, with a visit to Antioch almost certainly following shortly
afterwards.
Luke gives us more details about this period. In spite of his tendency to
be somewhat impressionistic in matters chronological, at least at this point
there are no major problems in reconciling Luke’s implied chronology
with that given by Paul himself. If we accept c. 33 as the probable date
of Paul’s conversion, it is reasonable to suppose that Greek-speaking Jews
from Jerusalem ‘scattered because of the persecution that took place over
Stephen’ (11.19) began their mission in Antioch in 36 or 37. At that point,
according to Luke, they began to speak to Gentiles as well (11.20), and
‘a great number became believers’. Luke states that ‘they told them the
good news of the Lord Jesus’, using the verbal form (eÉaggeliz»menoi),
not the noun, as is implied in many modern translations. The dramatic
developments in Antioch reached the ears of the church in Jerusalem, so
Barnabas was sent from Jerusalem to Antioch. He then went off to Tarsus
(in Cilicia), and brought Paul from Tarsus to Antioch (in Syria). Paul and
Barnabas lived in fellowship with the church in Antioch for a whole year
(Acts 11.22-6). And what was that year? It may well have been ad 39 or
40, just before Claudius became emperor following the murder of Gaius.
Luke underlines the significance of the time Paul and Barnabas spent in
Antioch by noting that it was there that followers of Jesus were first called
‘Christians’ (11.26).35
So on the basis both of Paul’s own letters and of Acts, it is reasonable to
suppose that it was among Greek-speaking Jews in Jerusalem, and perhaps
especially in Antioch between ad 37 and 40, that the gospel word group was
first used in a Christian context as God’s ‘glad tiding’ concerning Christ.

34 Martin Hengel and Anna Maria Schwemer, Paul between Damascus and Antioch (London: SCM,
1997), p. 27.
35 Riesner, Paul’s Early Period, p. 124, suggests 36/7 or perhaps 39/40 as the probable date.
24 Jesus and Gospel
Why might this particular date and geographical setting be significant?
The imperial cult first made a major impact on Jews in this region during
the reign of Gaius (Caligula), emperor from ad 37 until he was murdered
in 41.36 Writing very shortly afterwards, the Jewish philosopher Philo tells
us that the accession of Gaius was warmly welcomed: indeed, twice in his
De Legatione the verbal form eÉaggel©zesqai is used (§99 and §231). Gaius
is referred to as ‘saviour and benefactor, who would pour new streams of
blessings on Asia and Europe’ (¾ swtŸr kaª eÉerg”thv . . . phg‡v n”av
–pombržsein %s©a € te kaª EÉrÛph,€ De Legatione §22).
But within a couple of years relations between the emperor and his
Jewish subjects in both Alexandria and the whole of Judaea went sour.
Philo and Josephus give us different explanations for Gaius’ provocation
of Jews, but they agree that his promotion of the imperial cult was the
central issue (Philo, De Legatione §§184-348; Josephus, Ant. xviii.261-309).
Gaius was in fact the first Roman emperor to emphasize his own divinity:
he had cult statues from Greece shipped to Rome, where their heads were
replaced by models of his own.37 This is a clear case of modelling the
emperor on the gods. Augustus had been coy about making such claims
of divinity for himself, though he readily accepted ‘divine honours’ given
to him. Following ‘ethnic cleansing’ of Jews in Alexandria, and heightened
tension following an incident in Jamnia, Gaius rashly tried to have a statue
of himself erected in the temple in Jerusalem with the words: ‘Gaius, the
new Zeus made manifest’ (Di¼v –pijanoÓv N”ou, De Legatione §346).
Gaius instructed the governor of Syria, Petronius, to prepare the huge
statue. Petronius, who was based in Antioch, was well aware of Jewish
sensitivities. He knew full well that Gaius’ instructions had brought the
whole of Jewish Palestine and Syria to the verge of war.38 By deploying canny
delaying tactics Petronius managed to stave off the threat of war. It is hard to
believe that anyone from the whole area from Jerusalem to Antioch would
not have been fully aware of the political and religious crisis in ad 39/40.
Gaius was murdered in January 41, before matters came to a head.
Note that date. When Christian Greek-speaking Jews in Jerusalem and/or
Antioch were probably first starting to use the noun ‘gospel’ in the singular
to refer both to the act of proclamation of God’s glad tiding concerning
Jesus Christ and to its content, Gaius ordered his statue to be erected in

36 There had been earlier tensions. Twice a day sacrifices were offered in the temple for the emperor.
Pontius Pilate’s attempt to introduce troops into Jerusalem with the normal insignia bearing the effigy
of the emperor provoked outrage (Josephus, Bell. ii.169-74; Ant. 18.55-9). For detailed discussion see
Helen Bond, Pontius Pilate in History and Interpretation (Cambridge: Cambridge University Press,
1998), pp. 49–94.
37 Dio Cassius 59.28; Suetonius, Caligula 22. 38 Hengel and Schwemer, Paul, p. 181.
 Jesus and Gospel 25
the temple in Jerusalem. He was considered by many of his subjects to be
a ‘saviour and benefactor’. His accession had been hailed as ‘good news’,
and as marking the dawn of a new era, but his antics undermined that
acclamation. So, from a very early point indeed, Christian use of the gospel
word group may have formed part of a counter-story to the story associated
with the imperial cult.39
If so, it may be significant that, in the undisputed Pauline letters, neither
‘benefactor’ nor ‘saviour’ (with the exception of Phil. 3.20) is used of Jesus
Christ. Christians were proclaiming a rival gospel with some terminology
and themes shared with imperial gospels. But a line in the sand was drawn
at some key points – and avoidance of ‘saviour’ and ‘benefactor’ was one
such point. As we shall see, another was Christian insistence that there
was only one gospel, proclamation of God’s once-for-all provision of Jesus
Christ.
I have offered what I hope is a disciplined and responsible historical
reconstruction. However, the case I am advancing does not stand or fall
with my suggestions concerning Gaius and the years 39 and 40. My main
point is that the earliest Christian use of the phrase t¼ eÉagg”lion and
indeed of the verb eÉaggel©zesqai seems to have taken place between 37
and 40 in Jerusalem, or perhaps more probably in Antioch. Paul and his
co-workers may have taken this step themselves, or it may have been taken
by other Greek-speaking followers of Jesus. We cannot be certain.
The preceding paragraphs have opened up the possibility that Christians
borrowed the ‘gospel’ terminology from the imperial cult and filled it with
new content. The further evidence in the section that follows also suggests
that possibility, but even then it will fall short of proof. We may have
to concede that a quest for the origin of the distinctive Christian use of
the ‘gospel’ word group may not be able to locate the holy grail. But a
far more important point has already emerged. Early Christian use of this
word group seems to have developed alongside claims being made on behalf
of the Roman emperor. And it is that clue which will be followed up in
section 2.5.

2. 5 the gospel of jesus christ and the gospels

of the caesars
In the last two decades or so our knowledge of the imperial cult has increased
enormously. It can no longer be denied, as has happened in the past, that

39 On ‘story’ in Paul’s theology see especially B. Longenecker, ed., Narrative Dynamics in Paul (Louisville:
Westminster John Knox, 2002).
26 Jesus and Gospel
this is the backdrop against which Christians used the ‘gospel’ word group
in their own distinctive ways and made a number of their other claims
concerning Jesus Christ.40
Before I sketch out the main advances in recent scholarship, it will be
helpful if two near contemporary comments on the imperial cult are quoted.
Writing about the cult during Augustus’ reign, the historian Nicolaus of
Damascus is terse and to the point:
Because mankind addresses him thus (as Sebastos41 ) in accordance with their esti-
mation of his honour, they revere him with temples and sacrifices over islands and
continents, organized in cities and provinces, matching the greatness of his virtue
and repaying his benefactions towards them.42
In his lengthy, glowing panegyric on Augustus, Philo includes these
comments:
This is he who not only loosed but broke the chains which had shackled and pressed
so hard on the habitable world. This is he who exterminated wars . . . He was the
first and the greatest and the common benefactor . . . The whole habitable world
voted him no less than celestial honours. These are so well attested by temples,
gateways, vestibules, porticoes, that every city which contains magnificent works
new and old is surpassed in these by the beauty and magnitude of those appropriated
to Caesar and particularly in our own Alexandria.43
Two points are especially noteworthy in these two passages. Augustus is
universally revered, for his accession brought a new era of peace. Sacrifices
and the building of temples in his honour are the appropriate response of
his subjects to his magnanimous benefactions. Political loyalty lies behind
these comments, but there is also a strong note of religious devotion.
Scholarly study of the imperial cult has gathered pace in the last decade or
so.44 The evidence from literary sources, documentary sources (inscriptions
and papyri), archaeology, and numismatics is now being sifted with critical
40 See D. Georgi, ‘Die Stunde des Evangeliums Jesu und Cäsar’, in D. Georgi, M. Moxter, and H.-G.
Heimbrock, eds., Religion und Gestaltung der Zeit (Kampen: Kok Pharos, 1994), pp. 52–68.
41 S. R. F. Price notes that ‘the Latin “Augustus” was a title, implying divine favour, given to the first
emperor, whom we call Augustus, and employed by his successors. “Sebastos” is the Greek equivalent,
but has a stronger association with the display of religious reverence (eusebeia) to the emperor.’ Rituals
and Power: The Roman Imperial Cult in Asia Minor (Cambridge: Cambridge University Press, 1984),
p. 2 n. 1.
42 I have quoted Price’s translation, ibid., p. 1. For the Greek text, see F. Jacoby, Die Fragmente der
griechischen Historiker (Leiden: Brill, 1923–58), 90 f 125.
43 I have quoted F. H. Colson’s translation of De Legatione §§146–50 in Philo, LCL, Vol. x.
44 See especially S. J. Friesen, Twice Neokoros: Ephesus, Asia and the Cult of the Flavian Imperial Family
(Leiden: Brill, 1993); A. Small, ed., Subject and Ruler: The Cult of the Ruling Power in Classical
Antiquity. Papers presented at a conference held at the University of Alberta on April 13–15th, 1994, to
celebrate the 65th anniversary of Duncan Fishwick ( Journal of Roman Archaeology Supplementary
 Jesus and Gospel 27
rigour and with more careful attention to method than was often the case
in the past.45 Although it would be rash to claim that there is a scholarly
consensus, there would be considerable support for the following:
(1) The cult of the ruler, and especially of the emperor, was a central
element of ancient religious life. The Heidelberg ancient historian Géza
Alföldy goes even further: ‘Under the Roman Empire, from the time of
Augustus to that of Constantine, the cult of the emperor was, according
to the patterns of “religion” (not in a Christian sense but in the sense of
Roman religion) the most important type of worship.’46 It is no longer
acceptable to claim that the imperial cult was a Christian invention,47 or
that it was simply an expression of political loyalty.48
(2) Many aspects of the imperial cult can be traced back to Hellenistic
ruler cults, especially the emphasis on the importance of repaying the debts
of benefactions. However, with Augustus and the arrival of Empire there
are marked changes. ‘The Augustan decrees make explicit and elaborate
comparisons between the actions of the emperor and those of the gods.’49
In the Hellenistic period, ruler cults were usually city cults. These continued
in the Roman period, but in addition numerous cults were established by
the provincial assemblies.50
(3) The cult was not the preserve of the élite. All classes and groups in
cities and villages throughout the empire participated.51 However, evidence
from rural areas is sparse.52

Series 17 (1996)); A. Brent, The Imperial Cult and the Development of Church Order (Leiden: Brill,
1999); I. Gradel, Emperor Worship and Roman Religion (Oxford: Clarendon, 2002).
45 See Paul Zanker, The Power of Images in the Age of Augustus (Ann Arbor: University of Michigan
Press, 1990), p. 3: ‘My interest is . . . in the totality of images that a contemporary world would
have experienced . . . not only “works of art”, buildings, and poetic imagery, but also religious ritual,
clothing, state ceremony, the emperor’s conduct and forms of social intercourse.’
46 G. Alföldy, ‘Subject and Ruler, Subjects and Methods: An Attempt at a Conclusion’, in Small, Subject
and Ruler, p. 255. Cf. Price, Rituals and Power, p. 130: ‘The imperial cult . . . was probably the most
important cult in the province of Asia.’
47 G. Alföldy quotes the now generally rejected view of Kurt Latte (writing in 1958) that the imperial
cult was an invention of the church fathers. ‘Subject and Ruler’, p. 254. See also Zanker, Power of
Images, p. 299.
48 See Price, Rituals and Power, p. 55, for references to a number of scholars who have defended this
view. Note his insistence (p. 71) that ‘it is quite wrong to reduce the imperial cult to a pawn in a
game of diplomacy . . . It was not dreamed up simply to flatter the emperor.’ See also Zanker, Power
of Images, p. 299.
49 Price, Rituals and Power, p. 55. Contrast Stuhlmacher, Das paulinische Evangelium, p. 196: ‘Der
Kaiserkult gehörte, wenn man einmal so formulieren darf, mehr zur politisch-religiösen Engagement
der Vielen.’
50 Price, Rituals and Power, p. 56. So too F. Millar, ‘The Impact of Monarchy’, in F. Millar and E. Segal,
eds., Caesar Augustus: Seven Aspects (Oxford: Clarendon, 1984), p. 53.
51 Alföldy, ‘Subject and Ruler’, p. 255; Price, Rituals and Power, pp. 107–11.
52 Price, Rituals and Power, pp. 91–7.
28 Jesus and Gospel
(4) There was no slackening of interest in the cult under the successors
of Augustus; it continued (with numerous variations) until the end of the
third century. It is a mistake to suppose that early Christianity did not
feel its impact before the time of Domitian.53 Stephen Mitchell does not
exaggerate when he insists that public worship of the emperors was the
obstacle which stood in the way of the progress of Christianity, and was the
force which would have drawn new adherents back into conformity with
the prevailing paganism.54
(5) The imperial cult, ‘with its festivals, games, performances, processions
and public meals, must have been very attractive’.55 Justin Meggitt notes
that the cult seems to have been practised enthusiastically in private as well
as in public, though the material demonstrating this has generally been
neglected in studies to date and much more work remains to be done.56
(6) Ancient historians now insist that careful attention should be given
to chronology and to possible changes and developments, as well as to the
setting of the imperial cult in local contexts. I shall attempt to do this in
the following two sections of this chapter.

I shall now discuss some of the main examples of the use of the ‘gospel’
word group in the imperial cult. I shall begin with literary evidence from
Philo and Josephus before turning to inscriptions.
The Jewish philosopher Philo wrote at the time early Christian use began.
Philo does not use the noun at all in his extensive writings, but the verb is
found eleven times, usually with a non-religious sense, ‘to announce (good)
news’.57 In three passages, however, the context is especially important, for
here the language and ideology of the imperial cult are reflected.
At De Legatione §18 the recovery from sickness of the Emperor Gaius is
announced as good news, for he had at first been regarded as a ‘saviour and
benefactor’ who would ‘pour new streams of blessings on Asia and Europe’
(§22). Later in the same writing a reference to the speed with which good
news should be carried is part of the comparison of Gaius to the god Hermes

53 Pace Stuhlmacher, Das paulinische Evangelium, p. 197.

54 S. Mitchell, Anatolia: Land, Men and Gods in Asia Minor, Vol. ii (Oxford: Oxford University Press,
1993), p. 10.
55 Alföldy, ‘Subject and Ruler’, p. 255.
56 J. J. Meggitt, ‘Taking the Emperor’s Clothes Seriously: The New Testament and the Roman Emperor’,
in Christine E. Joynes, ed., The Quest for Wisdom: Essays in Honour of Philip Budd (Cambridge:
Orchard Academic, 2002), pp. 150–1.
57 For the details, see P. Borgen, K. Fuglseth, and R. Skarsten, The Philo Index: A Complete Greek Word
Index to the Writings of Philo of Alexandria (Grand Rapids: Eerdmans; Leiden: Brill, 2000).
 Jesus and Gospel 29
(§99). A report of the accession of Gaius is said to be good news for which
sacrifices were offered in the temple in Jerusalem (§231–2; see also §356).
Philo’s polemical and heavily ironical account of the reign of Gaius was
written shortly after the emperor died in ad 41, so the possible relevance of
these passages to earliest Christian use of the ‘gospel’ word group will be
obvious. We shall return to this point below.
Writing about three decades later, the Jewish historian Josephus uses the
verb a dozen times, usually with the sense ‘announce’, but in some passages
the announcement is of ‘good news’, especially news of a victory. In none
of these passages are there are any religious connotations. However, for
our present purposes, his three uses of the noun are of considerably more
interest.
Josephus records that at news of the accession of the new Emperor
Vespasian (ad 69) ‘every city kept festival for the good news (—Ûrtazen
eÉagg”lia) and offered sacrifices on his behalf ’ (Bell. iv.618). Similar ter-
minology is used at Bell. iv.656: the people in Rome celebrated (—Ûrtaze)
with one common festival the accession of Vespasian and the downfall of
Vitellius; on reaching Alexandria, Vespasian was greeted with this good
news (eÉagg”lia). The religious overtones are obvious. In both passages
eÉagg”lion is used in the plural.
The third passage is baffling. Josephus records that some leading citi-
zens of Jerusalem had sent a deputation to the Procurator Florus urg-
ing him to bring troops to the troubled city. ‘To Florus the news was a
wonderful godsend [or, preferably, ‘frightening’] (flÛrw„ m•n oÔn dein¼n
eÉagg”lion §n). Determined as he was to kindle the war, Florus gave the
emissaries no reply’ (Bell. ii.420). Here eÉagg”lion is unexpectedly used
in the singular, as it is without exception in NT writings, but without an
article. However, this passage is very difficult, if not impossible, to construe;
two textual variants confirm that early scribes were also as puzzled as we
remain.58
Gerhard Friedrich’s insistence that ‘neither in Philo nor in Josephus do
we find the same conception of the one who brings glad tidings as in
Deutero-Isaiah’ is undoubtedly correct.59 However, their use of our word
group in the context of the imperial cult points us towards paths worth

58 I have quoted Thackeray’s LCL translation of dein»n as ‘wonderful godsend’, but this is most unlikely
to be correct, for dein»v in Hellenistic Greek has the opposite sense, ‘causing or likely to cause fear’
(so BDAG). The phrase may be an oxymoron, ‘the terrifying good news’. See Stuhlmacher, Das
paulinische Evangelium, pp. 169–70 n. 2.
59 Gerhard Friedrich, art. eÉaggel©zomai in TDNT iii, p. 714.
30 Jesus and Gospel
following further, even though until recently they were considered by many
to be blind alleys.60
The literary evidence just discussed in Philo and Josephus is comple-
mented by evidence from an increasing number of inscriptions. The most
important is still the so-called Priene inscription, the first fragments of
which were published in 1899. Adolf Deissmann’s discussion of this in-
scription in his Licht vom Osten (1908) led to a flurry of interest in the
imperial cult. This book was quickly translated into English as Light from
the Ancient East (London: Hodder and Stoughton, 1910). It remains a clas-
sic, in spite of more recent discoveries and discussions.
Much has happened since Deissmann’s day, though little has filtered
through to the standard New Testament lexicons and handbooks. Many
more fragments of this inscription have been discovered; we now have
thirteen in all, from five cities in Asia Minor: Priene, Apamea, Maeonia,
Eumenia, and Dorylaeum.61 This inscription was displayed prominently
in Greek and in Latin in many more than these five places, not only in
the larger cities, but also in less populated areas. Only the well-known Res
Gestae Divi Augusti, the emperor’s own catalogue of his achievements for the
whole Roman Empire, had an even greater impact in the first century ad.
Copies of the Res Gestae in Latin (and often with a Greek translation or
paraphrase) were erected on stone blocks in the cities and towns of Asia
Minor, and probably also in Galatia at the instigation of the provincial
Assembly or koinon c. ad 19.62
The usual title, ‘Priene inscription’, is something of a misnomer. Priene,
which is about halfway between Ephesus and Miletus, happened to be the
place where the first discovery was made; the fragments found in the
other four cities are all less substantial, but that is sheer chance. When
the Ephesian elders travelled to meet with Paul at Miletus (Acts 20.15-17),
they may well have broken their journey at Priene.

60 Justin Meggitt notes that until recently NT scholars who have taken the figure of the Roman emperor
seriously have often found themselves the object of ridicule, and their interest regarded as, at best,
somewhat eccentric. ‘Taking the Emperor’s Clothes Seriously’, pp. 143–69.
61 The most comprehensive critical edition and discussion of all the fragments is Umberto Laffi, ‘Le
iscrizioni relative all’introduzione nel 9 a.C. del nuovo calendario della Provincia d’Asia’, Studi
Classici e Orientali 16 (1968) 5–98. See also Robert K. Sherk, Roman Documents from the Greek East:
Senatus Consulta and Epistula to the Age of Augustus (Baltimore: Johns Hopkins, 1969), pp. 328–37
for an edition of the Greek text, notes, bibliography, and brief discussion.
62 See Res Gestae Divi Augusti, ed. P. A. Brunt and J. M. Moore (Oxford: Oxford University Press, 1967).
See also S. Mitchell, ‘Galatia under Tiberius’, Chiron 16 (1986) 17–33. I owe the latter reference to T.
Witulski, Die Adressaten des Galaterbriefes. Untersuchungen zur Gemeinde von Antiochia ad Pisidiam
(Göttingen: Vandenhoeck & Ruprecht, 2000), p. 147.
 Jesus and Gospel 31
I prefer to refer to this inscription as the Calendar inscription, for it was
intended to encourage the replacement of the local lunar calendar with
the solar reckoning of the Julian calendar, as used in Rome. Changing the
calendar was a sensitive matter in antiquity. A recommendation to ‘do as
Rome does’ had to be handled with tact. So in about 9 bc the Proconsul
of Asia, Paulus Fabius Maximus, wrote a letter with his proposals to the
Provincial Assembly, the koinon responsible for emperor worship at the
provincial level. His suggestion is worded in such a way that it virtually
amounts to a directive.63 The Assembly duly responded with two decrees.
It formally approved the proposed change, and insisted that Paulus Fabius
Maximus should be honoured with a crown for suggesting that the Emperor
Augustus should be honoured by starting the new year on his birthday,
23 September. The letter and the decrees were inscribed many times over
on huge blocks of stone, which were then set up in cities all over Asia Minor.
Literacy levels were low in Asia Minor, so only a small percentage of
the population would have been able to read the lengthy inscription; even
fewer would have been able to appreciate the rhetorical flourishes. But
most people would have had a view about its message, for some cities did
not in fact fall for the rhetoric and failed to adopt the proposed calendar
reform.64
In this inscription the noun eÉagg”lion is used in the plural once; there
is almost certainly a second example in a damaged line. The context in
which this noun occurs is important. Here is the opening of Paulus Fabius
Maximus’ letter:
(It is hard to tell) whether the birthday of our most divine Caesar Augustus
(¡ toÓ qeiot†tou Ka©sarov gen”qliov ¡m”ra) spells more of joy or benefit, this
being a date that we could probably without fear of contradiction equate with the
beginning of all things (t¦i tän p†ntwn ˆrc¦i) . . . he restored stability, when
everything was collapsing and falling into disarray, and gave a new look to the
entire world that would have been most happy to accept its own ruin had not the
good and common fortune of all been born, Caesar Augustus. (lines 4-9)
‘The restoration of stability’, ‘a new look to the entire world’ – this
sounds like a press officer’s propaganda on behalf of her political masters.
‘The beginning of all things’ would have rung bells for the first Christians,
for they had a very different understanding of what constituted ‘the begin-
ning of all things’. The claim that Augustus was ‘most divine’ would have
caused many Christians to clench their teeth, for they claimed that it was

63 Sherk, Roman Documents, p. 334. 64 Price, Rituals and Power, p. 106.

32 Jesus and Gospel
appropriate to ascribe more than human qualities not to Augustus, but to
Jesus Christ.
Now part of the Provincial Assembly’s reply:
In her display of concern and generosity on our behalf, Providence (pr»noia), who
orders all our lives, has adorned our lives with the highest good, namely Augustus.
Providence has filled Augustus with divine power for the benefit of humanity, and
in her beneficence has granted us and those who will come after us [a Saviour] who
has made war to cease and who shall put everything [in peaceful order] . . . And
Caesar, [when he was manifest], transcended the expectations of [all who had
anticipated the good news], not only by surpassing the benefits conferred by his
predecessors but by leaving no expectation of surpassing him to those who would
come after him, with the result that the birthday of our god signalled the beginning
of good news for the world because of him (§rxen d• täi k»smwi tän di’aÉton
eÉangel©[wn ¡ gen”qliov ¡m]”ra toÓ qeoÓ) (lines 34–41).65
There is an unmistakable whiff of eschatology and of soteriology here. The
coming of the divine Augustus as ‘good news’ had been eagerly expected. He
came as saviour and benefactor, bringing benefits for all. He has brought
peace and will continue to do so. He was himself ‘the good news’. The
repetition in the reply to the original letter of the claim that the birthday
of Augustus was ‘the beginning of all things’ is especially striking.
The Calendar inscription is by no means an isolated example of the
ways in which Augustus was regarded. A lengthy but poorly preserved
decree opens with these words: ‘Since Emperor Caesar, son of god, god
Sebastos has by his benefactions to all men outdone even the Olympian
gods . . . ’.66 It is hard to see how Augustus could have been elevated any
higher.
Over two centuries later the accession of a new Roman emperor was still
considered to be ‘good news’. A papyrus letter written soon after ad 238
begins as follows: ‘Since I have become aware of the good news about
the proclamation as Caesar (of Gaius Julius Verus Maximus Augustus) . . .’
(–peª gn[Û]st[hv –gen»mhn toÓ] eÉangel[©o]u perª toÓ ˆnhgoreÓsqai
Ka©sara . . . ).67 His emperor father is described as ‘lord, most dear to the
gods’. This only non-Christian example of eÉagg”lion on papyrus uses
the noun in the singular. However, the key word is so damaged that we
cannot be completely certain that the singular should be read.

65 I have quoted with minor modifications F. W. Danker’s translation, Benefactor (St Louis, Miss.:
Clayton, 1982), pp. 216–17. The Greek is quoted from U. Laffi’s edition, ‘Le iscrizioni’.
66 I. Olympia 53, as quoted by Price, Rituals and Power, p. 55.
67 Sammelbuch griechischer Urkunden aus Aegypten 1 (1915) 421.2. I have quoted the text and translation
given by Horsley, New Documents, Vol. iii, p. 12. See also A. Deissmann, Light from the Ancient East
(London: Hodder and Stoughton, E. tr. 1910), pp. 371–2.
 Jesus and Gospel 33
At present this papyrus fragment and the baffling sentence from
Josephus, Bell. ii.420 quoted above are rare exceptions which prove the
rule: in Graeco-Roman writings and documents from the two centuries
before, and the three centuries after Christ, ‘good news’ is nearly always
found in the plural. In the small number of examples of eÉagg”lion in
the singular, the ‘secular’ setting of ‘good news’ is clear: there is no trace
of religious overtones.68 In sharp contrast, in Christian writings up to the
middle of the second century, the noun is always used in the singular. By
their choice of the singular, Christians were making a point. In the next
section of this chapter we shall consider when, where, and why they first
did so.
Equally important for our present purposes is the context in which the
word group is often used in non-Christian writings. Although the noun
is used in the plural without any religious connotations,69 the frequent
association of the word group with the imperial cult is clear. Examples
from the writings of Philo and Josephus in the first century were noted
above. The use of the noun in the Calendar inscription is but the tip of
an iceberg. G. H. R. Horsley has listed nine further examples of the use
in inscriptions of ‘good news’ in the plural (t‡ eÉagg”lia). He sums up
their contexts as follows: ‘The usage of the neuter plural noun is clear: it
refers to good news (often emanating from a monarch), such as news of
their victories or benefactions; and in particular, the word is employed of
the sacrifices celebrated on such an occasion. The occurrences are nearly all
Hellenistic in date.’70 So the roots are even deeper than the imperial cult
associated with Augustus and his successors.

2.5.1 Rivalry
It is now time to confront the key question. We have noted that the ‘gospel’
word group was very prominent in early Christian writings, and that the
use of the phrase t¼ eÉagg”lion, ‘the gospel’ in the absolute, is particularly
distinctive. The word group was used in non-religious contexts, but far more
striking is its use with religious overtones in connection with the imperial
cult that was almost all-pervasive in the decades in which Christianity first

68 So too Frankemölle, Evangelium, p. 89. Of the references given in BDAG, Ps.-Lucian, Asinus 26 and
Appian, Bella Civilia 3.92 are in the singular.
69 See Horsley, New Documents, Vol. iii, pp. 10–15. Horsley refers (inter alia) to Cicero’s use of eÉagg”lia
in three of his letters to Atticus: 2.3.1; 2.12.1; 13.40.1.
70 New Documents, Vol. iii, p. 13. In an e-mail dated 21 February 2000, Professor Horsley confirmed
that he and his colleagues have located further examples, which will be included in their forthcoming
new lexicon of the New Testament with documentary examples.
34 Jesus and Gospel
flourished. So what is the relationship between the uses of the gospel word
group in these two very different settings?
Early Christian addiction to the noun in the singular cannot readily be
explained either as a development of Scriptural usage or as influenced by
Jesus traditions, and even with the verb there is only limited continuity.
Wholesale borrowing from the imperial cult is equally implausible, for, as
we have seen, Christian use of the noun ‘gospel’ in the singular is almost
without contemporary precedent. Although there are some similarities in
terms of concepts and ideology, there are also very significant differences.
In the Graeco-Roman world of Paul’s day, ‘glad tidings’ were associated
regularly with the new hope, the dawn of a new era, the ‘good news’
brought about by the birth, the accession, or the return to health of a
Roman emperor. Hence there could be more than one set of ‘glad tidings’.
For Christians, on the other hand, the Gospel is God’s initiative, the good
news of God’s fulfilment of his plan and his purposes for humankind: its
focal point is Jesus Christ, God’s Son. At the heart of Paul’s theology was
the conviction that in the fulness of time God had sent forth his Son for
salvation, for redemption, for Jew and Gentile alike (Gal. 4.4-5). The life,
death, and resurrection of Jesus was God’s ‘once for all’ disclosure of ‘the one
glad tiding’. For Paul, his proclamation of ‘good news’ was not the ‘birthday’
of Christ which marked the dawn of the new era, as with Augustus, but
the crucifixion and resurrection of Christ. This was God’s good news.
I have been emphasizing the backdrop against which Paul’s ‘gospel’
proclamation would have been heard. For Paul himself there was another
backdrop, Scripture. He tells us so explicitly: the Gospel he gospelled to
the Corinthians (here he juxtaposes the noun and the verb) was in accor-
dance with the Scriptures (I Cor. 15.1-5). Paul insists that both the death
of Christ ‘for our sins’ and his raising to life were to be understood against
this backdrop. To ask which passages were in Paul’s mind when he wrote or
dictated these verses may be to ask the wrong question. Paul is concerned
with the general correspondence of the main themes of the Gospel with
Scripture.
So if – and, for the reasons given, it is a very big ‘if ’ – Paul, his predeces-
sors, and his co-workers ‘borrowed’ well-established current usage from the
all-pervasive imperial cult, they adapted it radically and filled it with distinc-
tively Christian content. And, of course, that content has deep Scriptural
roots. One set of religious themes was exchanged for another. Christian
proclamation of God’s provision of Jesus Christ first sounded out in a world
familiar with different ‘gospels’ (t‡ eÉagg”lia). Gospel proclamation did
not take place in isolation from the social, political, and religious culture
 Jesus and Gospel 35
of the time: it was regularly heard against the backdrop of the imperial
cult.
I do not think that we can be certain about the origin of Christian use
of the ‘gospel’ word group. At the end of section 2.4.1 I noted that a quest
for the origin of the distinctive Christian use of the ‘gospel’ word group
may not be able to locate the holy grail. The further evidence set out in
this section has confirmed that caution has to be the order of the day. But
what is clear is that there were rival ‘gospels’.71
What would have been ringing in the ears of those to whom Paul first
proclaimed God’s good news, and those who listened to his letters read
aloud? Not the ‘non-religious’ usage of the noun in the Greek Bible (and
perhaps not even the rich theological use of the verb in Deutero-Isaiah and
related passages), but the ‘religious’ usage of the word group in the imperial
cult which pervaded the cities in which Christianity first flourished. As
always, Gospel and culture are intertwined, and often somewhat at odds
with one another.
‘Gospel’ may have been adapted from its usage in the plural in the
imperial cult. Or it may have been adapted from its secular use, in which it
meant simply ‘good news’ without any religious connotations. But either
way it was modified radically, partly in the light of the Biblical usage of
the verb, and more particularly on the basis of early Christian convictions
concerning God’s salvific act through the death and resurrection of Christ.

2. 6 ‘gospel’ in the gal atian churches

I shall now take Paul’s letter to the Galatian churches as a test case. Were
Paul’s initial missionary proclamation and this letter itself heard by the first
recipients against the backdrop of the imperial cult? At first sight Galatians
seems to be most unpromising territory for such an inquiry. Scholarly
discussion has always focussed on a whole cluster of issues arising from
Paul’s dispute with the agitators whom Paul accused of wanting to pervert
the Gospel of Christ (Gal. 1.7). The social setting and the pre-conversion
religious beliefs of Gentiles in the Galatian churches have rarely received
sustained attention.
On the other hand, the gospel word group is particularly prominent in
this letter: together, the noun and the verb are used seven times. Only in
I Corinthians is the word group equally prominent – and that letter is about
71 I have chosen ‘rival’ deliberately. Justin Meggitt, ‘Taking the Emperor’s Clothes Seriously’, uses the
phrase ‘polemical rivalry’, but at least in Paul’s day it is difficult to find direct polemic being used
on either side.
36 Jesus and Gospel
three times as long as Galatians. The gospel word group seems to have been
used by Paul in his initial visit to the Galatian churches (1.7-8; 4.13).72 And
it is reasonable to infer that the agitators also used the noun ‘gospel’ (1.6-9).
Before we can consider the possible presence of the imperial cult in
the vicinity of the Galatian churches, we have to decide where they were
located. There has been no shortage of scholarly discussion of this old
chestnut. A decade or so ago it was common for writers on Galatians to state
their preference without bothering to discuss the merits of the alternative
destination.73 In recent years the whole question has been reopened in the
light of fresh considerations offered by ancient historians. In my judgement
the tide has turned decisively in favour of Pisidian Antioch, Iconium, Lystra,
and Derbe as the location of the churches of Galatia, i.e. cities along the Via
Sebaste in the southern part of the Roman province. Antioch, Iconium, and
Lystra were Roman colonies.74 It is no coincidence that Paul also sought
out Roman colonies at Philippi and Corinth.
Several fine very detailed studies have led to this emerging consensus.
They include three published in German;75 this is somewhat surprising,
as, in the past, German-language scholarship has generally supported the
so-called north Galatia or ethnic/territory theory; the Galatian churches
were thought to be in Ancyra, Pessinus, and Tavium.76
Perhaps the most significant voice in recent discussion has been the
ancient historian Stephen Mitchell, whose unrivalled knowledge of the
history and culture of Anatolia is now set out in two substantial volumes.
Mitchell is adamant: ‘there is virtually nothing to be said for the north
Galatian theory’.77 As far as I am aware, there has been no recent thorough
defence of the north Galatian theory – and one now seems most unlikely.78
72 M. Winger, ‘Act One: Paul Arrives in Galatia’, NTS 48 (2002) 548–67.
73 For example, H. D. Betz devotes only four paragraphs to this issue in his Hermeneia commentary
(Philadelphia: Fortress, 1979). In his Anchor Bible commentary (New York: Doubleday, 1998) J. L.
Martyn includes only three paragraphs. Both opt for the ‘north Galatia’ or ‘territory’ hypothesis.
74 For references to the literature, see C. Breytenbach, Paulus und Barnabas in der Provinz Galatien.
Studien zu Apostelgeschichte 13f.; 16, 6; 18, 23 und den Adressaten des Galaterbriefes (Leiden: Brill, 1996),
p. 1 n. 4.
75 Riesner, Paul’s Early Period (above, n. 33), esp. pp. 273–9; Breytenbach, Paulus und Barnabas; Witulksi,
Die Adressaten des Galaterbriefes.
76 If one does opt for the north Galatia theory, there is no shortage of evidence for the imperial cult
in Ancyra. The impressive temple of Roma and Augustus was already under construction by the
middle years of Augustus’ reign. The Latin text of the Res Gestae and a Greek paraphrase of it were
inscribed on the walls of the temple. In Pessinus, cultic worship of the emperors soon followed. See
Mitchell, Anatolia (above, n. 54), Vol. i, p. 103.
77 Ibid., Vol. ii, p. 3.
78 In his The Book of Acts in its Hellenistic Setting (Tübingen: Mohr, 1989), p. 307, C. J. Hemer doubted
whether it would be possible to make an adequate restatement of the north Galatia theory in a form
which attempted to make a balanced and representative use of epigraphical evidence.
 Jesus and Gospel 37
So the geographical details and ‘local colour’ in Luke’s account in Acts 13
and 14 of Paul’s so-called first missionary journey outside Syria and Palestine
may be taken seriously. Here Luke seems to have had access to a reliable
source, though one must allow for his own reshaping. Following contact
with the proconsul Sergius Paullus in Cyprus (and perhaps his conversion)
Paul and his companions sailed from Paphos to Perge and then travelled on
to the Roman colony at Pisidian Antioch on the Via Sebaste (Acts 13.4-14).79
Why did Paul not stay in Pamphylia, in the region of Perge and Attalia,
rather than undertake the arduous journey over the Taurus mountains to
Pisidian Antioch?
Perhaps Sergius Paullus (who eventually became consul at Rome c. ad 70)
persuaded Paul to make his hometown, Pisidian Antioch, his inland base.
This theory has won strong support on the grounds that the proconsul had
strong family ties in Antioch.80 Stephen Mitchell is even bold enough to
claim that it is ‘an elementary inference’ that Sergius Paullus advised or
encouraged Paul to make the trip up-country to Antioch.81 If so, then the
first people Paul would have met in the Roman colony would have been
members of the Roman élite, who would have been well acquainted with the
imperial cult.82 However, as this theory rests on little more than disciplined
imagination, we need firm evidence for the prominence of the imperial cult
in Pisidian Antioch before we can contemplate reading Galatians against
this background.
Pisidian Antioch was founded as a Roman colony in 25 bc, as was Lystra;
Iconium followed shortly afterwards. In Paul’s day Antioch was the second
city of the province of Galatia after Ancyra. Barbara Levick notes that
‘Antioch was a model of Rome, not only in its magistracies and priesthoods,
its deliberative body of substantial citizens, its vici and its seven hills, but
in the admixture of non-Italian elements within its walls’. She adds that
Antioch was an example of a Roman veteran settlement superimposed
upon the original town, ‘the result being a composite society, a bewildering
and contradictory variety of cultural and social phenomena, and a sharp
79 See especially Breytenbach, Paulus und Barnabas; P. Pilhofer, ‘Luke’s Knowledge of Antioch’, in T.
Drew Bear, M. Tashalan, and C. M. Thomas, eds., Actes du Ier Congrès International sur Antioche de
Pisidie (Lyons and Paris: Université Lumière-Lyon 2 and Diffusion de Boccard, 2000), pp. 77–84.
D. Campbell notes that Luke’s references to Perge and Attalia and Paul’s journey inland to Pisidian
Antioch on the Via Sebaste are ‘spot on’. ‘Paul in Pamphylia (Acts 13.13-14a; 14.24b–26): A Critical
Note’, NTS 46 (2000) 595–602.
80 Mitchell, Anatolia, Vol. ii, pp. 5–8; Breytenbach, Paulus und Barnabas, pp. 38–45.
81 S. Mitchell and M. Waelkens, eds., Pisidian Antioch: The Site and its Monuments (London,
Duckworth, 1998), p. 12.
82 S. Mitchell insists that Paul’s earliest mission to Asia Minor ‘was not aimed at low-status Anatolian
natives, still less at “foolish Galatians”, but at the Romanised provincial elite’. Ibid.
38 Jesus and Gospel
cleavage between upper and lower classes’.83 From Antioch Paul continued
to use the Via Sebaste en route to the Roman colonies of Iconium and
Lystra, before using another road, perhaps at that time unpaved, to Derbe.
He returned by the identical route to Antioch, thence to Perge and Attalia
(Acts 14.6, 8, 20-6).84
Paul used Antioch as his base; that makes sense as soon as one looks at
the province’s road network. Although archaeological excavation here has
been somewhat haphazard and is far from complete, we now have clear
evidence that the imperial cult was a prominent part of the city’s religious
life during Paul’s visits. In the very centre of the site stood a large temple
connected with a semicircular rock-cut area. Although attribution of the
temple has been debated ever since W. M. Ramsay’s excavations between
1912 and 1914, there is now general agreement that it is an imperial temple
from the time of Augustus or, at the very latest, Tiberius. Stephen Mitchell is
convinced that further stylistic analysis and a fresh reading of an inscription
on the triple-arched propylon (gateway) make it difficult to believe that the
building postdates Augustus’s death in ad 14. ‘The temple . . . was erected
for the cult of Augustus during his lifetime.’85
The inscription is especially important. It has now been reconstructed
from six fragments as follows:
IMP • CAES[ARI • DI]VI • [F•AVGVSTO • PONTI[F]ICI • M[AXIM]O
COS•X[III•TRIB]VN[ICIAE] • POTESTATIS • XXII • [IM]P • XIIII •
P[•P•]
Imp(eratori) Caes(ari) Divi f(ilio) Augusto pontifici maximo co(n)s(uli) XIII
tribuniciae potestatis XXII imp(eratori) XIIII p(ater) p(atriae)
For the emperor Caesar Augustus, son of a god, pontifex maximus, consul for the
13th time, with tribunician power for the 22nd time, imperator for the 14th time,
father of the country.
This reading gives a date of 2/1 bc. ‘The propylon, accordingly, was dedi-
cated to the living emperor Augustus, soon after he received the title pater
patriae on 5 February 2 bc.’86 Its decoration confirms the message inherent

83 Barbara Levick, Roman Colonies in Southern Asia Minor (Oxford: Oxford University Press, 1967),
pp. 190–1.
84 David French, ‘Acts and the Roman Roads of Asia Minor’, in D. W. J. Gill and C. Gempf, eds., The
Book of Acts in its Graeco-Roman Setting (Grand Rapids: Eerdmans, 1994), pp. 49–58.
85 For detailed discussion see chapter 5, ‘The Augustan Imperial Sanctuary’, in Mitchell and Waelkens,
Pisidian Antioch. Here p. 167.
86 For the text and translation see Mitchell and Waelkens, Pisidian Antioch, p. 147. The editio princeps has
not yet been published. However, in an e-mail dated 24 October 2002 Professor Mitchell confirmed
that in his opinion the provisional text given above is reliable.
 Jesus and Gospel 39
in the inscription: it was erected to honour Augustus and commemorate
the victories he had achieved and the peace these had brought to the Roman
world.87
This was also the message conveyed by the Res Gestae, Latin fragments of
which have been discovered at Antioch.88 In all probability the full text was
inscribed in ten columns on the inner faces of the two central piers of the
gateway. It opened as follows: ‘The achievements of the divine Augustus,
by which he brought the world under the empire of the Roman people . . .’
(‘Res gestae divi Augusti, quibus orbem terrarum imperio populi Romani
subiecet . . .’).89 The Res Gestae was published at the time of Augustus’
death in ad 14. ‘It can be taken as . . . his apologia for receiving his crowning
honour, state divinity, which he had so modestly (or prudently) rejected
throughout his lifetime.’90 A copy may well have been erected at Antioch
soon afterwards.
Thus there is cumulative evidence for the prominence of the imperial
cult in Antioch well before Paul’s day: the ground plan and decoration of
the temple, the decoration of the gateway, and the two inscriptions referred
to above. But what about Paul’s letter to the Galatian churches? Are there
passages in the text which may be references to the impact the imperial
cult may have made on the lives of Christians in Antioch? There are two
themes which may have been heard by Christians as a counter-story to the
story conveyed by the all-pervasive imperial cult, and two passages which
may possibly refer to that background.
(1) Two of Paul’s theological emphases in Galatians may have reminded
the Galatians that their convictions concerning Jesus Christ were at odds
with the beliefs and practices of the Roman religious and political world in
which they lived.
The gospel word group is more prominent in Galatians than in any
of Paul’s other letters. When Galatian Christians heard this word group
thirteen times in the opening sections of the letter, they may well have
recalled that in his initial preaching Paul insisted that there was one Gospel
of Jesus Christ, which was at odds with the ‘gospels’ associated with the
birth, accession, and health of the Roman emperors.

87 Mitchell and Waelkens, Pisidian Antioch, p. 164.

88 My graduate student Mr Justin Hardin has drawn my attention to the fact that no fragments in
Greek have yet been discovered in Antioch. I have benefited greatly from discussions with Mr Hardin
and from his M. Phil. thesis, ‘The Social and Religious Setting of Galatians’, Cambridge University,
2002.
89 See Brunt and Moore, Res Gestae Divi Augusti.
90 Gradel, Emperor Worship and Roman Religion (above, n. 44), p. 281.
40 Jesus and Gospel
At the centre of the imperial cult stood the recognition that Augustus was
a ‘son of god’, as were his successors. The inscription referred to in section 2.5
was set in bronze letters on the triple-arched gateway to the imperial temple
which dominated the city and the surroundings of Antioch. It would have
been a constant reminder of the religious status of the emperors. So Paul’s
counter-claim that Jesus, not the emperor, is God’s Son would have been
especially meaningful in that context.
And that is precisely the claim that lies at the heart of the Christology of
this letter. In the opening chapter Paul defines ‘the gospel’ (the noun is used
absolutely in 1.11) in terms of God’s disclosure of Jesus Christ as his Son. In
his positive statement about the origin of his gospel at the end of 1.12, Paul
insists that he received the gospel ‘through a revelation (di’ ˆpokalÅyewv)
of Jesus Christ’. This NRSV translation preserves the ambiguity of the
Greek, which can be construed either as ‘Jesus Christ’s disclosure of the
gospel’ or as ‘God’s disclosure of Jesus Christ as the content of the gospel’.
The latter is preferable, especially in view of Paul’s further comments in
1.15-16, which unpack his shorthand term ‘the gospel’ and emphasize (once
again) God’s initiative in the revelation or disclosure of his Son.
In his striking opening greeting Paul insists that Christ gave himself
for sins (Gal. 1.4), and at the climax of a sustained theological exposition in
2.15-20 Paul claims that the Son of God loved him and gave himself for
him (Gal. 2.20). These would have been jarring notes for those accustomed
to hear that the emperor as divi filius, a son of god, had brought many
benefactions to the world, including peace and general well-being. Self-
giving love on behalf of others, even to the point of death, was not part of
the imperial repertoire.
Similarly in what I take to be the theological heart of this letter: 4.1-7. In
the fulness of time God sent his Son: this is God’s good news concerning
Jesus Christ. In the Calendar inscription the beginning of good news for
the world is Providence’s provision of Augustus as a saviour who has made
wars cease. But there the partial parallelism stops abruptly. For in imperial
cult ideology there is no note of redemption or of the bestowal of sonship
by adoption on those committed to faith in the Son of God.
Now, I am not suggesting that either the content of Paul’s gospel or a
Son of God Christology was borrowed or even strongly influenced by the
language and ideology of the imperial cult. But, in the light of the evidence
I have set out, we do well to bear in mind the religious and political world
in which Paul’s initial proclamation and his letter to the Galatian churches
were heard. The Gospel of Jesus Christ and the gospels of the Caesars were
rivals.
 Jesus and Gospel 41
(2) If the general point just made is accepted, then it is worth considering
whether there are passages in Paul’s letter to the Galatian churches which
may refer to possible pressures to practise the imperial cult. Gal. 4. 8-10 is
a candidate.
Formerly, when you did not know God, you were enslaved to beings that by nature
are not gods. 9. Now, however, that you have come to know God, or rather be
known by God, how can you turn back again to the weak and beggarly elemental
spirits (stoice±a)? 10. How can you want to be enslaved to them again? You are
observing special days, and months, and seasons, and years.
The consensus reading of this passage, which I have followed in the
past, runs as follows.91 Before they became Christians, the Galatians
were enslaved to ‘beings that by nature are not gods’, i.e. to idols (cf. I
Cor. 8.5 and 12.2). So they must have been Gentiles. Now as believers they
have come to know, i.e. to experience, God’s Spirit (cf. 3.1-5 and 4.6). But
what are the special ‘days, and months, and seasons, and years’ which the
Galatians now want to observe closely, even as Christians? I have assumed
that the influence of the agitators may be detected here. They have been
encouraging Galatian Christians to observe the Jewish sabbath (‘days’) and
festivals (‘seasons’). Observance of ‘months’ probably refers to observance
of the new moon which marked the beginning of each month. Precisely
what is meant by ‘years’ is uncertain.
However, there is an alternative reading which merits consideration.
H. D. Betz notes that the cultic activities described in verse 10, ‘observing
special days, and months, and seasons, and years’, are not typical of Judaism
(including Jewish Christianity). He might have added that, unlike Col. 2.16,
this verse does not refer explicitly to Jewish observance of sabbaths, new
moons, or festivals. Betz insists that ‘the description is typical and part of a
literary topos well-known in antiquity. It portrays the Galatians as conform-
ing to the religious character of the deisida©mwn (“religiously scrupulous”
or even “superstitious”).’92 However, Betz fails to explain satisfactorily why
Paul has used this particular ‘non-Jewish’ phraseology to refer to the threat
posed by Jewish Christian agitators. Nonetheless, his comments may point
us in the right direction.
Thomas Witulski has recently gone much further down this path. He
notes that the absence of any explicit reference to the sabbath, the most
important day of all for Jews, rules out a reference to Jewish observances

91 See G. N. Stanton, ‘Galatians’, in J. Barton and J. Muddiman, eds., The Oxford Bible Commentary
(Oxford: Oxford University Press, 2001), pp. 1152–65.
92 Betz, Galatians (above, n. 73), pp. 217–18.
42 Jesus and Gospel
here. Equally problematic for the consensus reading is the reference to
‘years’, for Jewish celebration of ‘special years’ was unknown in diaspora
Judaism.93 Witulski then presents a wealth of evidence to underpin his
claim that the Galatian Christians were facing local pressures to participate
in the imperial cult.94
In principle, it is highly likely that Galatian Christians did face strong
social pressures, even though these may be more difficult to uncover from
Paul’s letter than the theological issues at stake between Paul and the agi-
tators. Stephen Mitchell concludes his discussion of the imperial cult in
Anatolia with these words:
One cannot avoid the impression that the obstacle which stood in the way of the
progress of Christianity, and the force which would have drawn new adherents
back to conformity with the prevailing paganism, was the public worship of the
emperors . . . It was not a change of heart that might win a Christian convert back to
paganism, but the overwhelming pressure to conform imposed by the institutions
of his city and the activities of his neighbours.95
Witulski notes that several celebrations of ‘special days’ were associated
with Augustus. ‘Months’ may refer to the introduction of the new pattern of
months referred to in the Calendar inscription, with its strong encourage-
ment to begin the new year on the birthday of Augustus on 23 September.
Careful observance of ‘seasons’ is taken to refer to festive celebrations in
the context of the imperial cult, some of which extended over several days.
‘Years’ may reflect Augustus’ reference in the Res Gestae to the decision of
the Roman senate that every fifth year ‘vows should be undertaken for my
[i.e. Augustus’] health by the consuls and the priests. In fulfilment of these
vows games have frequently been celebrated.’ In this paragraph Augustus
then notes that all citizens are included in these acts of devotion: ‘individ-
ually and on behalf of their towns they offer . . . prayers at all the shrines
for my health’.96 I have summarized only part of the considerable evidence
Witulski has amassed in building a cumulative case.
If this line of interpretation is followed, there is an obvious problem. How
are social pressures to conform to the imperial cult linked to the particular
concerns of the agitators in the Galatian churches over circumcision and
observance of the law? Witulski’s own solution smacks of desperation. He
claims that two letters Paul wrote to the Galatians have been combined
to form what became the canonical Galatians. The two letters originally
had very different settings: in the earlier letter Paul attacks the agitators,

93 Witulski, Die Adressaten des Galaterbriefes, pp. 155–6. 94 Ibid., pp. 158–68.
95 Mitchell, Anatolia, Vol. ii, p. 10. 96 Brunt and Moore, Res Gestae Divi Augusti, ch. 9, p. 23.
 Jesus and Gospel 43
in the other (at least partly now in Gal. 4.8-20) he encourages resistance
to pressures to observe the imperial cult. The first letter was written before
the arrival of the imperial cult in Pisidian Antioch, the latter shortly after
its arrival about ad 50. However, this dating (and the whole theory) is
undermined by the more recent evidence sketched above, which indicates
that the imperial cult arrived well before ad 50. And Gal. 4.8-20 is not the
only part of Galatians which may have been heard against the background
of the imperial cult.
(3) If Gal. 4.8-10 reflects the enormous social pressures to observe the
imperial cult, then in 6.12-13, the opening verses of the carefully composed
finale to the letter, we may perhaps be given a more specific explanation
of the social pressures at work in the Galatian churches. Bruce Winter has
recently proposed a solution that merits careful consideration. He notes
that, immediately after Paul personally begins to pen the postscript to the
Galatians, in his own hand, the apostle sums up his key concern in quasi-
legal terminology in 6.12-13:
As many as wish to secure ‘good’ status by means of flesh, i.e. circumcision, [these
people] compel you to be circumcised only in order that they may not be persecuted
for the cross of Christ. 13. For those who themselves are circumcised do not keep
the law, but only wish you to be circumcised in order that they may have confidence
by means of your flesh, i.e. your circumcision.
Winter claims that avoidance of the imperial cult is connected to the
central issue in Galatians. Gentile Christians were being encouraged by
Jewish Christian agitators to ‘look Jewish’, i.e. to undergo circumcision
and to keep the law as the means of circumventing their obligation to the
imperial cult. ‘Jewish Christians formulated this response to an extremely
difficult civic obligation, for their own self-preservation and that of the
Christian community was seen to be at stake.’ The agitators believed that
‘undergoing circumcision and keeping the law was one way of convincing
the authorities that Christianity was part of a religio licita, for in Galatia
these were cultural hallmarks of the one group that was recognized as being
exempted from worship of the emperors.’97
This is certainly a provocative fresh reading of Galatians, which should
not be dismissed without further ado. Winter correctly understands that
the term religio licita is no more than a convenient way of referring to the

97 B. W. Winter, ‘The Imperial Cult and Early Christians in Roman Galatia (Acts xiii 13-50 and
Galatians vi 11-18’, in T. Drew-Bear, M. Tashalan, and C. M. Thomas, eds., Actes du Ier Congrès
International sur Antioche de Pisidie (Université Lumière-Lyon 2 and Diffusion de Boccard, 2002),
pp. 67–75.
44 Jesus and Gospel
special status Jews enjoyed: at this time there was no formal charter which
underpinned that status. Even though we do not have specific evidence
from the middle of the first century of the status of Jews in Roman eyes
in the colony at Antioch,98 it is highly likely that the Galatian Christians
were being encouraged to accept circumcision in order to take advantage
of the respect Jews enjoyed in society at large. John Barclay notes that ‘by
becoming proselytes the Galatians could hope to identify themselves with
the local synagogues and thus hold at least a more understandable and
recognizable place in society’.99
However, it is difficult to take a further step and suppose that the pres-
sures on the Jewish Christian agitators were quite specific and involved the
imperial cult. The final clause in Gal. 6.12 is one of the most difficult in
Galatians: Paul claims that the agitators are trying to compel circumcision
‘only that they may not be persecuted for the cross of Christ’ (m»non ¬na
tä€ staurä€ toÓ CristoÓ mŸ diÛkwntai). The agitators are wanting to
avoid ‘persecution’ by compelling Gentiles to be circumcised. That much
is clear. But who was responsible for the threat of persecution? Paul does
not identify the alleged persecutors: he shelters them behind his use of the
passive verb.
Some have suggested that too much weight should not be attached to this
clause. Perhaps it is a ‘throw-away’ line, penned by Paul himself in a state
of fury. Perhaps we can envisage strong arm-twisting from some quarter
or other, but persecution? On the other hand, at 6.12 Paul uses the same
verb (diÛkw) as at Gal. 1.13, where he refers to his own ‘pre-conversion’
persecution of the church. So it is natural to suppose literal persecution.
Plenty of candidates are on offer. The ‘false brothers’ of Gal. 2. 4-6, or ‘hard-
line’ Jews, whether local or new arrivals in Antioch, are often thought to
have been the culprits.
Winter’s theory offers a very different scenario. The Jewish Christian
agitators were under severe pressure to participate in the imperial cult and
thus prove that they merited the respect and toleration accorded to members
of the Jewish community. Hence they urged Gentile converts in Antioch
to ‘look Jewish’ and thus enjoy that same standing in society. In support
of his theory Winter refers to Acts 13.50 and 14.12. ‘The leading men of
98 See especially J. M. G. Barclay’s judicious appraisal of the limited evidence for the social status of
Jews in the province of Asia, Jews in the Mediterranean Diaspora (Edinburgh: T & T Clark, 1996),
pp. 259–81. He notes (p. 279) that it is frustrating that evidence for the social setting of Asian Jews in
the first century is so paltry. He does not discuss Galatia. See also Paul Trebilco, Jewish Communities
in Asia Minor (Cambridge: Cambridge University Press, 1991), pp. 167–85.
99 J. M. G. Barclay, Obeying the Truth: A Study of Paul’s Ethics in Galatians (Edinburgh: T & T Clark,
1988), p. 60.
 Jesus and Gospel 45
the city’ were involved in the ejection of Paul and Barnabas from Antioch.
Antagonism towards the new movement is likely to have continued. Hence,
on his return visit to Antioch, Paul warns the Christians there that they
must expect to face continuing troubles. There is no reason to dismiss this
evidence for Gentile opposition to Christians in Antioch out of hand, but
there are no hints in Acts 13.50 and 14.12 that ‘the leading men of the city’
were enforcing observance of the imperial cult on Christians.
Winter acknowledges that we do not have evidence from the period
that explains how Gentile Christians might have claimed exemption from
participation in the imperial cult. This is a weak link in his argument,
but not quite an Achilles’ heel. Our knowledge of the imperial cult in
Antioch in the middle of the first century has increased enormously in
recent decades, but it is still limited. It is to be hoped that much-needed
further archaeological work will enhance that knowledge. So it would be
prudent to accept that at present we do not have evidence which would
either confirm or undermine the envisaged scenario.100
(4) There is one further passage to be considered. Luke includes a
lengthy account of Paul’s preaching in the synagogue in Pisidian Antioch in
Acts 13.14-41, one of the great set-piece speeches in the first half of Acts.
Although it is extremely difficult to disentangle Luke’s own colouring
and shaping from any sources he may have used in the speeches in Acts,
one should take seriously the fact that Luke sets this speech in Pisidian
Antioch. For Luke regularly takes pains to shape the speeches in Acts to
fit the local context. The preceding speech made by Peter to the Roman
centurion Cornelius in Caesarea is a good example (Acts 10.34-43), as is
Paul’s Areopagus speech (Acts 17.22-31). So, even if we are sceptical about
the extent to which Luke has used sources in the speech in Acts 13.14-41,
some of its themes may have been intended to fit the context envisaged, as
is clearly the case in the speech set in Lystra (Acts 14.15-17).
If we bear in mind that in Pisidian Antioch the synagogue would have
stood in the shadow of the huge imperial temple, then there may be a
ready explanation for two surprises Luke springs in this speech. The first
Christological note sounded is that God has brought to Israel a Saviour,
Jesus, as he promised (13.23); ‘to us, this message of salvation has been sent’
(13.26; see also 13.47). ‘Saviour’ (swtžr) is found in only one other passage
in Acts (5.31) and is rare in the NT. ‘Salvation’ (swthr©a) is used sparingly
by Luke (only here and at Acts 4.12; 7.25; 16.17; 27.34).

100 My comments on Gal. 6.12 have benefited from discussion with my colleague Dr James Carleton
Paget, but he is not responsible for their formulation.
46 Jesus and Gospel
Is it a coincidence that ‘saviour’ and ‘salvation’ were prominent in the
terminology of the imperial cult, that the imperial cult was prominent in
Antioch, and that both terms were included in Luke’s account of Paul’s
speech in this city?101 As we noted above, according to Philo, the accession
of the Emperor Gaius was announced as good news, for he had at first
been regarded as a ‘saviour and benefactor’. Swtžr is particularly common
in inscriptions referring to the Emperor Claudius.102 So the listeners to
the speech in Luke’s day may well have been reminded that Jesus, not
the emperor or any other person or god, was Saviour, God’s provision of
salvation. However, it is unlikely that this theme goes back to Paul himself,
for, with the exception of Phil. 3.20, in the undisputed Pauline letters
swtžr is not used at all. The only possible link between the speech and
Galatians is at Acts 13.38-9, which might be taken as a partial paraphrase of
Gal. 2.16.
A Son of God Christology is also found only once in Acts up to this point
(at 9.20), and is not repeated.103 At the climax of the speech Luke’s Paul
proclaims the good news (eÉaggeliz»meqa, 13.32) that Jesus is the Son of
God. God has fulfilled his promises: Jesus is the one of whom it is written
in Psalm 2, ‘You are my Son; today I have begotten you’, the fulfilment of
Psalm 2.7 and II Sam. 7.12 (13.23, 32-3). As we have seen, both ‘son of god’
and ‘announcement of good news’ are often associated with the imperial
cult. So once again the readers of Acts are being reminded that Jesus, not
the emperor, is to be acclaimed as Son of God. If we take seriously the
setting of the speech in Acts 13.14-41 in Pisidian Antioch, we are forced to
consider a reading which has been largely overlooked.

In the preceding paragraphs of this section I have suggested that Paul’s initial
proclamation of the Gospel, as well as his letter to the Galatian churches,
would have been heard in the Roman colonies of Pisidian Antioch, Iconium,
and Lystra against the backdrop of the all-pervasive imperial cult. This claim
is strengthened by the use of the ‘gospel’ word group in both settings, but it
is not dependent on this verbal link. For the whole ideology of Providence’s
provision of the emperor as the supreme benefactor, son of god and saviour
was in ‘rivalry’ with Paul’s claim concerning God’s provision in grace /
benefaction of Jesus Christ as God’s Son.

101 For details, see Peter Oakes, Philippians: From People to Letter (Cambridge: Cambridge University
Press, 2001), pp. 138–47. See also W. Foerster, art. swtžr, TDNT vii, pp. 1010–12.
102 See Oakes, Philippians, p. 140 for details.
103 The confessional response of the Ethiopian eunuch, ‘I believe that Jesus Christ is the Son of God’
(Acts 8.37), is clearly a later addition to the text.
 Jesus and Gospel 47
2.7 ‘gospel’ in thessalonica and philippi
There is even stronger evidence for this contention from some of Paul’s
other letters.104 I shall limit myself to two examples, one from the beginning
and one from the close of Paul’s letter-writing ministry: the Thessalonian
correspondence, and Philippians. Thessalonica was strongly Romanized,
and Philippi was a Roman colony.
J. R. Harrison has recently argued that ‘a pivotal claim of early first cen-
tury propaganda – that providence would never provide a better Saviour
than Augustus – increasingly faced challenge at Thessaloniki and else-
where’.105 The accusation that Paul and Silas acted ‘against the decrees
of Caesar, saying there is another king Jesus’ (Acts 17.7) is the obvious start-
ing point. Harrison lists no fewer than six words from the Thessalonian
correspondence which are common to the imperial and early Christian
eschatology: e«ržnh (‘peace’: I Thess. 1.1; 5.3, 23); –pij†neia (‘appearance’:
2 Thess. 2.8); –lp©v (‘hope’: I Thess. 1.3; 2.19; 4.13; 5.8; 2 Thess. 2.16);
eÉagg”lion (‘good news’: I Thess. 1.5; 2.2, 4, 8, 9; 3.2; II Thess. 1.8;
2.14); swthr©a (‘salvation’: I Thess. 5.8, 9; II Thess. 2.13); c†ra (‘joy’:
I Thess. 1.6; 2.19, 20; 3.9). He claims that Paul countered and subverted the
aggressive influence of the eschatology of the imperial gospel by proclaim-
ing the eschatological hope of the risen and reigning heavenly Lord.106 In
my judgement, his revival of the theory that at I Thess. 5.3 Paul prophesies
the destruction of the proponents of the imperial pax et securitas is parti-
cularly strong. Equally plausible is the claim that the phrases ‘the hope of
salvation’ and ‘salvation through our Lord Jesus Christ’ in I Thess. 5.8-9
would have evoked imperial associations.
Given that in the Thessalonian letters the most frequently used of the
six words listed in the preceding paragraph is eÉagg”lion, it is surprising
that Harrison does not discuss this term. For our current preoccupation
with the gospel word group, two points are especially important. First, in I
Thessalonians, probably Paul’s earliest letter, Paul uses ‘the word’ (¾ l»gov)
and ‘the gospel’ (t¼ eÉagg”lion) almost synonymously (I Thess. 1.6, 8 and
2.13; see also II Thess 3.1). In the opening thanksgiving in I Thessalonians,
the Spirit is associated closely with the reception both of ‘the gospel’ and
of ‘the word’ (1.5-6); the terms could easily be interchanged in these verses.
104 See R. A. Horsley, Paul and Politics: Ekklesia, Imperium, Interpretation (Harrisburg, Pa.: Trinity
Press International, 2000). For Romans see N. T. Wright, ‘A Fresh perspective on Paul?’, BJRL 83
(2001) 21–39. For the imperial cult in Corinth see J. K. Chow, Patronage and Power: A Study of Social
Networks in Corinth (Sheffield: JSOT Press, 1992).
105 J. R. Harrison, ‘Paul and the Imperial Gospel at Thessaloniki’, JSNT 25 (2002) 71–96.
106 Ibid., p. 92.
48 Jesus and Gospel
Both are ‘shorthand’ phrases whose content could be readily filled out by
the Thessalonian Christians. In some cases Paul himself expands both terms
slightly by adding ‘of God’ or ‘of Christ’ to ‘the gospel’ (I Thess. 2.2, 8, 9; 3.2)
and ‘of the Lord’ or ‘of God’ to ‘the word’ (I Thess 1.8; 2.13; II Thess 3.1).
I shall comment further on the ‘shorthand’ character of both terms in
section 2.8.
In I Thessalonians, the apostle takes considerable pains to emphasize that
the Gospel or the word is God’s – it is from God, and it is about what God
has done (2.2, 8, 9, 13). The Gospel and the word make their appeal not
by means of fancy rhetoric, but on the basis of God’s power or Spirit (1.5).
The Gospel is God’s glad tiding about Christ (3.2). The repeated emphasis
in this letter on the Gospel as God’s good news is striking. Why does Paul
need to do this when he assumes that the Thessalonians will be able to
unpack his shorthand? Surely there is at least an implicit subversion of the
imperial gospel of Providence’s provision of the emperor for salvation, for
‘peace and security’. This is how Paul’s words would have been heard in
Thessalonica, whether or not that was part of Paul’s intention.
And so briefly to Paul’s letter to Christians in the Roman colony, Philippi.
Phil. 3.20 has often been read against the background of the imperial cult:
‘our citizenship is in heaven, and it is from there that we are expecting a
Saviour (swtžr), the Lord Jesus Christ’. The so-called Christ hymn in
Phil. 2.6-11, with its climax in the confession that Jesus Christ is kÅriov,
has been read similarly, but less often. Peter Oakes has recently built on
earlier readings along these lines by showing that some of the key words and
phrases in these passages (especially swtžr and kÅriov) were prominent in
the imperial cult. His cumulative case for reading the whole of Phil. 2.6-11
as an extended comparison between Christ and the emperor is provocative
and may not win the day. Nonetheless, in the light of what has been said
above, surely his main point is plausible: this is how many of the Philippians
would have heard this passage.107
Peter Oakes provides plenty of literary, inscriptional, and numismatic
evidence to back up his case, but hardly any of it is local to Philippi. Perhaps
this is why he accepts that this cult is unlikely to have been the most pressing
of the issues faced by the Philippians.108 In effect, Oakes is offering a timely
warning against the dangers of ‘mirror-reading’, i.e. of assuming too readily
that every phrase in Paul’s letters was either crafted by the apostle or heard
by the Philippians in the light of the imperial cult.
In spite of this caveat, there is a striking feature of this letter, not men-
tioned by Peter Oakes, which may be related indirectly to the imperial cult.
107 Oakes, Philippians, pp. 147–74. 108 Ibid., p. 137.
 Jesus and Gospel 49
The phrase ‘the gospel’ is more prominent in this letter than in any other
early Christian writing. In the opening chapter Paul uses the phrase t¼
eÉagg”lion six times, five of which are in absolute construction (Phil. 1.5,
7, 12, 16, 27), i.e. without any explanatory phrase. Phil. 1.27 is the excep-
tion that proves the rule, for here we find ‘the gospel of Christ’. In three
further passages in this letter the phrase is used absolutely (Phil. 2.22; 4.3,
15). The verb is not used at all. Once again ‘the word’ (¾ l»gov) is used in
the absolute, and synonymously with ‘the gospel’ (Phil. 1.14).109
Paul does mention that ‘the gospel’ has become known throughout the
whole Praetorian Guard (Phil. 1.13). Members of the imperial guard would
certainly have known the rival ‘gospels’ concerning Providence’s provision
of emperors. And so too, we may surmise, Christians in the Roman colony
at Philippi.
Paul’s repeated use of ‘the gospel’ and his use of ‘the word’ (Phil 1.14)
as shorthand terms is striking. The Philippian Christians are able fill out
their content on the basis of Paul’s initial proclamation in their city and
his continuing concern for them. Indeed, unlike Galatians, from this letter
alone it is difficult to set out the content of ‘the gospel’. What we have in
Philippians, as in Paul’s other letters, is use of a cluster of shorthand terms
well known to the recipients – terms which are filled with distinctively
Christian content and which are used in ways which are out of kilter with
wider use in non-Christian first-century settings. I shall take this point
further in the next section of this chapter.

2. 8 ‘gospel’ as part of an early christian sociolect

In an important article on the rhetorical functions of ‘the gospel’ in the
Corinthian correspondence, Margaret Mitchell has built on Wayne Meeks’s
earlier observation that Paul’s letters are packed with ‘extremely compact
formulas’. She notes that the phrase t¼ eÉagg”lion serves as a ‘superab-
breviation’ of the whole narrative sequence of events which describe God’s
salvific acts on behalf of humankind. The phrase functions as a title which
both characterizes the full contents of ‘the gospel’ and interprets its meaning
for the hearer.110

109 There is strong manuscript support for ‘the word of God’ (t¼n l»gon toÓ qeoÓ); indeed, this
reading was adopted in NA 26. F and G read t¼n l»gon kur©ou. NA 27 follows the lead of ∏46
and omits toÓ qeoÓ. This is not always a wise move, given that ∏46, our earliest MS of Paul’s
letters, is riddled with minor slips. In this case, however, the shorter reading is preferable; later
scribes failed to appreciate Paul’s shorthand use of ‘the word’.
110 M. M. Mitchell, ‘Rhetorical Shorthand in Pauline Argumentation: The Functions of “the Gospel”
in the Corinthian Correspondence’, in L. A. Jervis and P. Richardson, eds., Gospel in Paul, FS
50 Jesus and Gospel
This is a helpful observation. I have already borrowed Margaret Mitchell’s
phrase, ‘Paul’s shorthand’. I fully endorse her main conclusion. ‘Paul’s punc-
tuated abbreviations unite his readers with himself and one another in a
common bond of shared language and assumptions, a task central to the
formation of ecclesial self-identity and social cohesion, and at the same time
allow for elegant economy of expression in the new literary creations, the
letters.’111 However, I am not yet persuaded that we need to follow Margaret
Mitchell’s path in order to reach that conclusion. She claims that Paul is
using effectively three forms of ancient rhetorical shorthand, described in
rhetorical theory as ‘brevity’, ‘synecdoche’, and ‘metaphor’. However, the
extent to which Paul draws on ancient rhetorical theory is much disputed.112
I am much more inclined to view Paul’s use of shorthand as a classic ex-
ample of the use of language patterns as identity markers, and here we can
turn to sociolinguists for assistance.
The discipline of sociolinguistics is barely four decades old. Whereas
Durkheim and Weber, the fathers of sociology, paid little attention to lan-
guage, sociolinguists are now busy making up for lost time. A major theme
in their research is the function of language as a means of group formation.
In 1966 William Labov, one of the pioneers, observed the linguistic acts
of identity made by those who wished to be recognized as natives of the
summer resort Martha’s Vineyard.113 Lesley Milroy has made similar ob-
servations about language used on the streets of Belfast.114 J. K. Chambers,
one of the leaders in the field, emphasizes that ‘the underlying cause of
sociological differences . . . is the human instinct to establish and maintain
social identity’. People have a profound need to show that they belong
somewhere, and to define themselves by the use of language. ‘We must also
mark ourselves as belonging to the territory, and one of the most convincing
markers is speaking like the people who live there.’115
Religious, political, ethnic, and other social groups (even teenagers) do
develop their own ‘insider’ terminology, often by adapting the vocabulary

R. N. Longenecker (Sheffield: JSOT Press, 1994), pp. 63–88; here 63–4; W. Meeks, The First Urban
Christians: The Social World of the Apostle Paul (New Haven: Yale, 1983), p. 93. See also Betz,
Galatians, pp. 27–8, who lists a number of Pauline ‘theological abbreviations’, but not (surprisingly)
t¼ eÉagg”lion.
111 Mitchell, ‘Rhetorical Shorthand’, p. 88.
112 See especially P. H. Kern, Rhetoric and Galatians (Cambridge: Cambridge University Press, 1998).
113 For a helpful survey of the development of the discipline, see R. B. Le Page, ‘The Evolution of
a Sociolinguistic Theory of Language’, in Florian Coulmas, ed., The Handbook of Sociolinguistics
(Oxford: Blackwell, 1997), pp. 15–32.
114 Lesley Milroy, Language and Social Networks, 2nd edn (Oxford: Blackwell, 1987).
115 J. K. Chambers, Sociolinguistic Theory (Oxford: Blackwell, 1995), p. 250.
 Jesus and Gospel 51
used by ‘outsiders’ so radically that the language of the ‘in-group’ is virtually
incomprehensible beyond its own boundaries.
Biblical scholars should not be surprised to learn that there is a strong link
between language and identity, and that even a single feature of language
suffices to identify someone’s membership of a given group. In Judg. 12.5-6
we read that, when any escaping Ephraimite wished to cross the Jordan,
the men of Gilead would ask, ‘Are you an Ephraimite?’, and if he said,
‘No’, they would retort, ‘Say “Shibboleth”.’ He would say ‘Sibboleth’, thus
disclosing his identity and sealing his fate.116
I have argued above that in the middle of the first century the difference
between ‘gospel’ in the singular and ‘gospels’ in the plural was just as
significant. ‘Gospel’ in the singular was developed and used in the very
earliest post-Easter communities to summarize the good news of God’s
once for all provision of Jesus Christ. This linguistic pattern was honed in
the teeth of the rival ‘gospels’ of imperial propaganda. It quickly became
part of the shared assumptions of Paul, his co-workers, and the communities
he established. In Paul’s early letters, the Thessalonian correspondence and
Galatians, an explanatory phrase is often tacked on to ‘the gospel’; absolute
use of the phrase is uncommon. In what is probably Paul’s final letter,
Philippians, ‘the gospel’ is nearly always used absolutely. By then there was
no need to add explanatory phrases, for ‘the gospel’ was set in stone as a
key building block in a distinctively Christian language pattern.
Early Christian use of the gospel word group, and especially the noun,
is one of the most prominent tips of a large iceberg. Here we have a term
which had very different associations for insiders and outsiders. There were
others. I have mentioned more than once that ‘the word’ was used by Paul as
a synonym for ‘the gospel’, and that it was used similarly as shorthand, often
in the absolute. In fact, it is arguable that ‘the word’ as a compact summary
of the Christian message was used even more widely in the first century
than ‘the gospel’. It was used, for example, by the author of Hebrews and
by Luke, both of whom shunned the phrase ‘the gospel’.
Other early Christian near-synonyms for ‘the gospel’, such as ‘the faith’
(¡ p©stiv, Gal. 1.23), ‘proclamation’ (t¼ kžrugma), ‘the message’ (¡ ˆkož,
e.g. Gal. 3.5), and also ¡ ˆggel©a (I John 1.5), and ‘witness’ (t¼ martÅrion)
functioned similarly. They were not unknown to ‘outsiders’, but to ‘insiders’
they had a distinctive nuance.
The term c†riv (‘grace’/‘benefaction’) is a first cousin of t¼ eÉagg”lion.
J. R. Harrison’s fine study of c†riv complements some of the key points

116 See further, A. Tabouret-Keller, ‘Language and Identity’, in Coulmas, Sociolinguistics, pp. 315–26.
52 Jesus and Gospel
made in this chapter.117 He notes that, in sharp contrast to the LXX, Paul
chose c†riv over against ›leov. In Paul’s day c†riv was normally used in
the plural, and often to denote the beneficent dispensations of the Roman
emperor. Paul always uses the term in the singular, and fills it with Biblical
and distinctively Christian content. ‘The c†ritev of Augustus had acquired
soteriological, eschatological, and cosmological status within his own life-
time throughout the Graeco-Roman world. The grace of the Caesars would
remain a continuing refrain. It was precisely in this context that Paul
announced God’s reign of grace through Christ . . . The paradox is that
God accords the status of righteousness to believers through a dishonoured
Benefactor.’118
I hope that I have said enough to make out my case. The very ear-
liest Christians developed their own ‘in-house’ language patterns, partly
on the basis of Scripture, especially the Septuagint, partly in the light of
their distinctive Christian convictions, but partly by way of modifying con-
temporary ‘street’ language. In this way they developed their own ‘social
dialect’, and in turn this was very influential on their self-understanding,
and their worldview.
They did not, however, develop a wholly new language. If they had done
so, obviously evangelism would have been impossible. Here we have a
continuing dilemma for Christian theology, one which is underlined by the
fact that today there is an increasingly wide gap between the ways insiders
and outsiders understand the term ‘gospel’. How are Christians to develop
language which expresses Christian convictions and yet is intelligible to all
and sundry? This is part of an even grander theme, the relationship of the
Christian Gospel to culture.

2. 9 from oral procl amation to writ t en narrat ive

Sociolinguists would surely be fascinated by one of the most startling de-
velopments in the linguistic usage of the first Christians. In the earliest
post-Easter decades t¼ eÉagg”lion refers to oral proclamation of the sig-
nificance of the death and resurrection of Jesus, not to a written account
of the story of Jesus. There is general agreement that by no later than the
second half of the second century t¼ eÉagg”lion was used to refer to a
‘gospel-book’.
117 J. R. Harrison, ‘Paul’s Language of Grace (C†riv) in its Graeco-Roman Context’, Ph.D. thesis,
Macquarie University, 1996, to be published in the WUNT series by J. C. B. Mohr, Tübingen. See
also his ‘Paul, Eschatology and the Augustan Age of Grace’, TB 50 (1999) 79–91.
118 Harrison, ‘Paul’s Language of Grace’, p. 201 and p. 210.
 Jesus and Gospel 53
There were two phases to this very considerable linguistic development,
the first less radical than the second. The evangelist Mark envisages that
‘the gospel’ includes an account of the teaching of Jesus and narratives of
his actions. Mark is expanding the semantic field associated with earlier
Christian usage of the noun, especially in Pauline circles. The actions and
words of Jesus are now integrated fully into Christian proclamation of God’s
glad tiding concerning the death and resurrection of Jesus Christ. Although
this is often singled out as Mark’s most important theological achievement,
I do not think that Mark was as innovative as most suppose. For I have
long been persuaded that oral proclamation of ‘Christ crucified and raised’
by Paul and his co-workers would hardly have made sense without at least
a sketch of the story of Jesus.119 Pauline specialists are now more willing
to acknowledge that there is a narrative sub-structure to Paul’s theology.120
Nonetheless, the term ‘gospel’ has undergone considerable development
between Paul and Mark.
An even more considerable development in early Christian usage of
‘gospel’ came with the use of this term to refer to a written account of the
teaching and actions of Jesus. Who was responsible for this extension of
the linguistic usage of the very first Christians? Opinions differ markedly.
Was it Mark, or Marcion, or should we look elsewhere?
Luke and John rule themselves out of contention immediately. Luke
does not use the noun at all in his account of the life of Jesus. No one has
ever explained satisfactorily why he did not do so. The Fourth evangelist
uses neither the noun nor the verb. This is no surprise, for he transposes a
great deal of earlier Christian vocabulary into a different key.
So who first used ‘gospel’ to refer to a writing?121 In tackling questions of
this kind it is often helpful to work backwards from the point at which a
new development is well attested. From the end of the second century and
the first half of the third century there is clear evidence from papyri, the Old
Latin and the Coptic versions, and from Irenaeus, Clement of Alexandria,
and Tertullian, that the titles ‘the Gospel according to Matthew, Mark etc.’
(eÉagg”lion kat‡ Maqqa±on, M†rkon ktl.) were widely used.122 In the
middle of the second century Justin, I Apol. 66.3, provides the first example
119 See G. N. Stanton, Jesus of Nazareth in New Testament Preaching (Cambridge: Cambridge University
Press, 1974), pp. 86–116.
120 See Longenecker, Narrative Dynamics in Paul.
121 The remaining paragraphs in section 2.9 are an abbreviated version of part of my contribution to the
Festschrift in honour of Franz Neirynck: ‘Matthew: b©blov, eÉagg”lion, or b©ov?’, F. Van Segbroeck
et al., eds., The Four Gospels 1992, Vol. ii (Leuven: Leuven University Press, 1992), pp. 1190–5.
122 See M. Hengel, ‘The Titles of the Gospels and the Gospel of Mark’, in his Studies in the Gospel
of Mark (London: SCM, 1985), p. 66f. Hengel argues convincingly against the widely held view
54 Jesus and Gospel
of the plural eÉagg”lia to refer to ‘gospel books’. Justin states explicitly
that the ˆpomnhmoneÅmata tän ˆpost»lwn (a phrase he uses thirteen
times in Dialogue 100-7 and once more in I Apol. 67.3-4) are eÉagg”lia.123
About a decade earlier Marcion almost certainly referred to his version
of Luke as ‘the gospel’.124 H. Koester has claimed that Marcion’s concept
of a written gospel was a ‘revolutionary innovation’ which Justin took
over.125 However, Koester partly undermines his own case by making two
concessions. (1) Even before Marcion’s day, liturgical readings from the
written gospels may well have been introduced as ‘the gospel’.126 (2) Koester
accepts Hengel’s claim that the canonical gospels must have circulated under
the name of a specific author from the very beginning, but he denies that
the original titles were identical with the later ‘Gospel according to . . .’.127
It is almost inconceivable that the name of the author would have been
attached to copies of the gospels without a title of some kind.128 But what
title would have been used, if not t¼ eÉagg”lion?129
There is in fact evidence that eÉagg”lion was beginning to be used
to refer to a writing in the decades before Marcion. (a) II Clement 8.5
introduces one of his quotations of words of Jesus with ‘the Lord says in the
Gospel’ (l”gei g†r ¾ kÅriov –n tä€ eÉaggel©w„). The writing concerned is
probably Luke’s Gospel, 16.10a and 11.
(b) In two of his letters, Ignatius (c. ad 110) uses ‘the Gospel’ (t¼
eÉagg”lion) to refer to proclamation of the cross and resurrection of
Jesus. At Philad. 9.2, t¼ eÉagg”lion includes the coming (parous©a here

(which is reflected in the titles of the four gospels in the 26th (and also the 27th) edition of the
Nestle–Aland text) that a shorter form (kat‡ Maqqa±on etc.) of the later titles was original.
123 See Chapter 4.
124 See A. von Harnack, Marcion: Das Evangelium vom fremden Gott (Leipzig: J. C. Hinrichs, 1921),
pp. 165–6*.
125 H. Koester, ‘From the Kerygma-Gospel to Written Gospels’, NTS 35 (1989) 361–81, esp. 377–81; the
phrase quoted is used on p. 381. While at first sight it does seem unlikely that Justin would adopt
a usage introduced for the first time by his arch-rival Marcion, this is not impossible. Rivals often
influence one another to a much greater extent than they are aware.
126 Ibid., p. 381 n. 1. M. Hengel had made the same point in some detail earlier. See his ‘Titles’,
pp. 74–81.
127 ‘Kerygma-Gospel’, p. 373 n. 2.
128 See M. Hengel, ‘Titles’, p. 65. At a very early point Papias refers to the names of the writers of
the gospels without using eÉagg”lion, but he is commenting on the origin of the writings – not
referring to the title of a manuscript.
129 Koester, ‘Kerygma-Gospel’, p. 374 suggests that what Papias says about both Matthew and Mark
reveals that these ‘gospels’ had incipits which were similar to those still preserved in the gospels
from the Nag Hammadi Library. But this does not take us further forward. According to Koester’s
own survey in his Ancient Christian Gospels: Their History and Development (London: SCM and
Philadelphia: Trinity, 1990), pp. 20–3, most of the incipits and colophons of the Nag Hammadi
‘gospels’ are either missing or clearly later additions to the manuscripts. Matthew and Mark are
most unlikely ever to have been referred to as ‘Secret Sayings’, or ‘Secret Book’!
 Jesus and Gospel 55
undoubtedly refers to the life) of Jesus Christ, as well as his suffering and
resurrection. This suggests that a writing rather than oral proclamation is
being referred to, but the latter interpretation is possible.130
In the letter to the Smyrnaeans, however, there is less room for doubt.
The letter opens with a credal summary of Christological convictions which
refers to the baptism of Jesus by John, ‘so that all righteousness might be
fulfilled by him’ (¬na plhrwq¦€ pŽsa dikaiosÅnh Ëpì aÉtoÓ). The verbal
agreement with Matt. 3.15, a verse which bears the stamp of Matthew’s
redactional hand, is sufficiently close to persuade most scholars that Ignatius
is here quoting Matthew’s Gospel.131
In the same letter (5.1) Ignatius notes that neither the prophecies, nor
the law of Moses, nor t¼ eÉagg”lion has persuaded his opponents. The
juxtaposition of t¼ eÉagg”lion with Scriptural writings strongly suggests
that, in this letter at least, Ignatius is referring to a writing – most probably
Matthew’s Gospel, which he has quoted just a few paragraphs earlier. A
similar juxtaposition of t¼ eÉagg”lion with Scriptural writings is found
two paragraphs later at 7.2. Here Ignatius urges his readers to ‘pay at-
tention to the prophets and especially to the gospel (tä€ eÉaggel©w„) in
which the Passion has been made clear to us and the resurrection has been
accomplished’.132
(c) In Didache 8.2; 11.3; 15.3-4f. there are four references to t¼
eÉagg”lion. The first passage introduces a quotation of the Lord’s prayer
with wording which is close to Matt. 6.9-13. The other two passages allude
to Matthew’s versions of sayings of Jesus. Since few scholars doubt that
the Didache is dependent on Matthew, these references confirm that t¼
eÉagg”lion was used to refer to a gospel writing, almost certainly Matthew,
some decades before Marcion.133
If Marcion was not the first to use the term eÉagg”lion to refer to
a written account of the story of Jesus, was Mark the innovator, some

130 W. R. Schoedel insists (but in my view without compelling arguments) that Ignatius always uses t¼
eÉagg”lion to refer to oral proclamation, not to a writing. See his Ignatius of Antioch, Hermeneia
(Philadelphia: Fortress, 1985), p. 208 n. 6; p. 234.
131 See W. Köhler, Die Rezeption des Matthäusevangeliums in der Zeit vor Irenäus, WUNT 24 (Tübingen:
Mohr, 1987); O. Knoch, ‘Kenntnis und Verwendung des Matthäus-Evangeliums bei den Apos-
tolischen Vätern’, in L. Schenke, ed., Studien zum Matthäusevangeliums, FS W. Pesch (Stuttgart:
Katholisches Bibelwerk, 1988), pp. 167–8.
132 I have quoted M. W. Holmes’s light revision of J. B. Lightfoot’s The Apostolic Fathers: M. W.
Holmes, ed., The Apostolic Fathers: Greek Texts and English Translations, 2nd edn (Grand Rapids:
Baker, 1992).
133 Koester, ‘Kerygma-Gospel’, pp. 371f., assigns these passages rather arbitrarily to the final redaction
of the Didache, which he dates to the end of the second century. This late dating and his denial of
the dependence of the Didache on Matthew are accepted by few other scholars.
56 Jesus and Gospel
seventy years earlier? For Mark, t¼ eÉagg”lion is the proclaimed message
about Jesus Christ. His usage is close to Paul’s, even though Mark, unlike
Paul, sets out a narrative of the teaching and actions of Jesus as a ‘sermon’
on, or expression of, the Gospel.134 This is how Mark’s opening line, ‘the
beginning of the gospel of Jesus Christ’ (ˆrcŸ toÓ eÉaggel©ou ’IhsoÓ
CristoÓ), should be interpreted. Whether toÓ eÉaggel©ou is taken as a
subjective or as an objective genitive, Mark 1.1 refers to ‘proclamation’, not
a ‘written report’.135 While it is true that Mark’s development of Paul’s use
of t¼ eÉagg”lion paves the way for later reference to the written story of
the life of Jesus as t¼ eÉagg”lion, Mark did not take that step himself.
Matthew, however, did so. In order to show this, we must examine
closely the differences between the use of ‘gospel’ in Matthew and in
Mark. Matthew omits five of Mark’s uses of eÉagg”lion (Mark 1.1; 1.14
and 15; 8.35; 10.29) and expands the other two (Mark 13.10 and 14.9). With
the exception of Mark 1.1, in all these passages Matthew is following Mark
closely, so the omissions are striking. They have led M. Hengel to conclude
that Matthew uses t¼ eÉagg”lion ‘only in a markedly reduced sense’.136
Since in his very much longer gospel Matthew retains only two of Mark’s
seven uses of the noun and adds it to only two other passages, this seems to
be a reasonable conclusion. On closer inspection, however, it is clear that
Matthew’s use of this key word is an important new development.
First of all we must account for Matthew’s omission of five of Mark’s
uses of eÉagg”lion. Mark 1.1 is omitted as it would have been an inap-
propriate introduction to the genealogy Matthew included. Mark 1.14 and
15 are omitted as part of Matthew’s concern to bring out clearly the close
correspondence between the proclamation of John the Baptist (Matt. 3.2),
Jesus (4.17), and the disciples (and followers of Jesus in his own day) (10.7):
they all proclaim the coming of the kingdom of heaven.
Mark 8.35; 10.29 are omitted because the phrases ‘for the sake of the
gospel’ and ‘for the sake of Jesus himself’ are almost synonymous, and
point to a post-Easter setting. Matthew, however, saw the earthly Jesus, his
teaching and actions, as ‘gospel’, and so omitted the Marcan passages.

134 W. Marxsen’s discussion of Mark’s use of t¼ eÉagg”lion has been influential. See his Mark the
Evangelist, especially pp. 126–38.
135 W. Marxsen suggests that it is almost accidental that something in the way of report also appears,
Mark the Evangelist, p. 131. R. Guelich argues strongly that 1.1 is not a title for the whole work
since syntactically it must be linked with the sentence which follows. See his Mark, Word Biblical
Commentary (Dallas: Word, 1989), p. 9, and, more fully, ‘The Gospel Genre’, in P. Stuhlmacher,
ed., Das Evangelium und die Evangelien (Tübingen: Mohr, 1983), pp. 183–219, esp. pp. 204–8.
136 Hengel, ‘Titles’, p. 83.
 Jesus and Gospel 57
Unlike Luke, Matthew is not averse to the noun. He uses it in two key
passages, 4.23 and 9.35, to summarize the proclamation of Jesus as ‘the
gospel of the kingdom’ (t¼ eÉagg”lion t¦v basile©av). From the context
it is clear that in both passages he has the teaching of Jesus in mind. These
passages are central pillars in Matthew’s construction of the first half of
his gospel: between these pillars he places the Sermon on the Mount in
chapters 5–7 and his cycle of miracle traditions in chapters 8–9.
The two passages in which Matthew retains Mark’s use of eÉagg”lion
are equally revealing. In 24.14 Matthew expands Mark’s absolute use of the
noun at 13.10 to ‘this gospel of the kingdom’ (toÓto t¼ eÉagg”lion t¦v
basile©av). In 26.13 Matthew has ‘this gospel’ (t¼ eÉagg”lion toÓto);
once again Mark’s absolute use of the noun at 14.9 is modified. In both
passages the addition of toÓto is very striking. The redactional phrase ‘the
word of the kingdom’ (¾ l»gov t¦v basile©av) at Matt. 13.19 is clearly
closely related to 24.14 and 26.13.
What is ‘this gospel of the kingdom’ (or ‘word of the kingdom’, 13.19) with
which the readers of Matthew are to confront the whole world? No defi-
nition or explanation is given in any of these three passages, but Matt. 26.13
provides an important clue. This verse, which refers to the woman’s act of
anointing the head of Jesus with costly oil, shows that for Matthew ‘this
gospel of the kingdom’ includes not only the teaching of Jesus, but also
accounts of his actions.137
As J. D. Kingsbury notes, the evangelist simply assumes that his readers
will know what ‘this gospel’ is on the basis of their acquaintance with his
written document.138 We may conclude with Kingsbury that the phrase
‘this gospel of the kingdom’ is Matthew’s own capsule-summary of his
work.139
Matthew probably did not provide a title for his writing, but he intended
his full account of the teaching and actions of Jesus to set out for his readers

137 Marxsen, Mark the Evangelist, p. 124, claims that for Matthew the ‘gospel is of a piece with his
speech complexes’, but 26.13 surely rules that out. On p. 141 he concedes that there is ‘a kernel
of truth’ in the hypothesis that Matthew’s addition of toÓto is intended to set up a connection
between the Gospel as such and the book of Matthew.
138 J. D. Kingsbury, Matthew: Structure, Christology, Kingdom (London: SPCK, 1975), p. 130, and also
p. 163.
139 Ibid., 131. Kingsbury refers to several earlier writers who have supported the same conclusion. See
now also W. Schenk, Die Sprache des Matthäus (Göttingen, Vandenhoeck & Ruprecht, 1987), p. 265.
U. Luz, Matthäus, p. 249 is sympathetic to this view, but suggests that a direct link is not yet drawn.
In Ancient Gospels, p. 11, n. 4, H. Koester rather rashly claims that all modern commentaries agree
that ‘this gospel’ at Matt. 26.13 cannot refer to Matthew’s Gospel, thus overlooking the commentaries
of J. Schniewind (on Matt. 24.14) and W. Grundmann. For a different view, see R. H. Gundry,
ëEUAGGELION: How Soon a Book?’, JBL 115 (1996) 321–5.
58 Jesus and Gospel
the content of ‘this gospel of the kingdom’. As soon as an individual or a
community had access to more than one narrative account of the life of
Jesus, it would have been necessary to distinguish between them by means
of a title, especially in the context of readings at worship.140 That first
happened as soon as Matthew had completed his writing, for the evangelist
(and perhaps some of the communities to which he wrote) then had two
accounts of the story of Jesus: his own, and Mark’s.
We may conclude, then, that long before Marcion’s day ‘gospel’ was used
for a written account of the story of Jesus. But, since Mark’s usage is closer
to Paul’s and is very different from Matthew’s, Mark is not the ‘radical
innovator’.141 Need we look any further than the evangelist Matthew? By
his insistence that the teaching and actions of Jesus are ‘gospel’ (4.23 and
9.35) and by his addition of ‘this’ to Mark’s ‘the gospel’ in 24. 14 and 26.13
(toÓto t¼ eÉagg”lion), he indicates clearly to his readers that there is a
close correspondence between the ‘gospel of Jesus’ and ‘the gospel’ which is
to be proclaimed in his own day. He also emphasizes that his own writing
is ‘a gospel’.142
We should not suppose that, once t¼ eÉagg”lion began to be used for a
written account of the life and teaching of Jesus, it ceased to be used to refer
to Christian oral proclamation. That double usage continues to this day.
The preceding paragraphs confirm that in early Christian writings it is
often difficult to determine whether t¼ eÉagg”lion refers to oral proclama-
tion or to a writing. Oral traditions did not disappear the moment Mark
wrote. As we shall see in Chapters 3 and 4, oral traditions and written
gospels continued to exist side by side until at least the end of the second
century.
Language is rarely static. The term t¼ eÉagg”lion is a prime example.
In Paul’s day it referred to the one oral Gospel of God’s provision of Jesus
Christ, in contrast to Providence’s repeatable ‘gospels’ of the provision of
Roman emperors. By the end of the second century Irenaeus drops his

140 This point is made most impressively by M. Hengel in ‘Titles’; see his section entitled ‘The Practical
Necessity of the Titles’, pp. 74–81.
141 M. Hengel, who sees Mark as the innovator, does not discuss the differences between Matthew
and Mark in ‘Titles’. I accept Hengel’s point that Mark may have been referred to as ‘the gospel’
as soon as it began to be used in readings in worship. If so, then this usage would have influenced
Matthew. However, the first explicit evidence for the use of t¼ eÉagg”lion for a writing is found
in Matthew’s Gospel, not Mark’s.
142 For a very different view, see H. Frankemölle, Jahwebund und Kirche Christi (Münster: Aschendorff,
1974), who claims that Matthew is a literary work, a Buch der Geschichte modelled on Jewish history
writing such as Deuteronomy and Chronicles. This claim does not do justice to the importance for
Matthew of the OT prophetic writings and of Mark’s Gospel.
 Jesus and Gospel 59
guard and refers to four written gospels, even though he emphasizes that
there is one Gospel according to four individual evangelists.

2 .10 co nclusions
This discussion of the origin and use of t¼ eÉagg”lion in earliest Chris-
tianity has necessarily ranged far and wide. I do not intend to weary readers
with summaries of my main conclusions. They can be found at the end of
most of the individual sections of this chapter. Nonetheless, several points
cry out for brief further comment.
I have suggested that, very soon after Easter, Greek-speaking Jewish
followers of Jesus began to use the ‘gospel’ word group to refer to procla-
mation of God’s one glad tiding concerning Jesus Christ. The content of
that proclamation would have been heard by many with the language of
the imperial cult ringing in their ears, whether or not that was intended by
Christian missionaries and teachers who used this terminology. Of course,
the imperial cult did not provide the first Christians with their central theo-
logical themes. Those themes were shaped by Scripture and by convictions
concerning the life, death, and resurrection of Christ. Although there were
some superficial similarities between the two forms of ‘good news’, early
Christian proclamation of the Gospel was distinctive, and ultimately sub-
versive of its rival. The key question was: whose good news? Providence’s
provision of Caesars and their benefactions, or God’s once for all provision
of grace through ‘a dishonoured Benefactor, Jesus Christ?’143
As we shall see in Chapter 6, long before the downfall of Jesus, his own
proclamation of God’s glad tiding was deemed by some to be subversive.
The key question was: was Jesus a demon-possessed magician and a false
prophet? Or was Jesus proclaiming in word and action God’s good news to
the poor as a messianic prophet?
Contemporary Christian theological statements regularly assume that
a straight line of continuity concerning ‘gospel’ can be traced from Jesus
to the post-Easter church. This has long been the traditional view. It was
defended in 1908 by James Denney in an influential book whose title aptly
sums up his argument: Jesus and the Gospel: Christianity Justified in the Mind
of Christ.144 It is expressed very forcefully in the Second Vatican Council’s
Constitution on Divine Revelation ‘Dei Verbum’, chapter 2 §7 (1965). After
143 I owe the striking phrase ‘dishonoured benefactor’ to Harrison, ‘Paul’s Language of Grace’, p. 210.
144 (London: Hodder & Stoughton, 1908). In spite of his title, James Denney includes only one
paragraph on the term ‘gospel’. Somewhat surprisingly, he does not hesitate to accept that Mark
uses the term ‘gospel’ in the sense of the apostolic church (p. 57).
60 Jesus and Gospel
noting that Christ had ‘commissioned the Apostles to preach to all men that
Gospel which is the source of all saving truth and moral teaching, and to
impart to them heavenly gifts’, the statement continues: ‘This Gospel had
been promised in former times through the prophets, and Christ Himself
had fulfilled it and promulgated it with His lips.’
My own line of continuity between Jesus’ own proclamation and post-
Easter proclamation of him is less neat and tidy. It is curved in places, and
dotted in others. Nonetheless, there is continuity at some points. Jesus of
Nazareth was more than a proclaimer of good news. Key phrases in 4Q521
strongly suggest that in his use of Isaiah 61 Jesus was making an indirect
messianic claim: he was himself part of God’s good news.
It is often overlooked that the evangelists have less hesitation than modern
scholars in drawing that line of continuity. As we have seen, Mark does this
through his use of t¼ eÉagg”lion. This is confirmed by the interpretation
of the parable of the sower (4.13-20). Here we are told that the seed sown
by the sower is ‘the word’, and from the context this can only be a reference
to the Gospel; it is clearly implied that the sower of the word or the Gospel
is Jesus himself. In fact ‘the word’ and ‘the gospel’ are synonymous here,
as in several passages in Paul’s letters, and also in Matthew, as Matt. 13.19
confirms.
Matthew’s line of continuity is even thicker. In his first two uses of t¼
eÉagg”lion he has in mind the teaching of Jesus (Matt. 4.23 and 9.35).
In his only two other uses of the phrase (Matt. 24.14 and 26.13), he envis-
ages that his own writing will serve as ‘gospel’ proclaimed ‘throughout the
world’.
Luke does not use the phrase t¼ eÉagg”lion. Nonetheless, he also stresses
that the proclamation of Jesus and post-Easter proclamation are identical.
He does this by means of the verb ‘to proclaim good news’ (eÉaggel©zesqai)
and the phrases ‘the word’ and ‘the word of God’. Just as Jesus proclaims
God’s good news (Luke 4.18, 43; 7.22; 8.1; 9.6; 16.16; 20.1), numerous
passages in Acts report that the apostles did likewise. At Luke 5.1 we read
that the crowd pressed in on Jesus ‘to hear the word of God’. The phrase
echoes the words of Jesus two verses earlier (4.43): ‘I must proclaim the good
news of the kingdom of God’ (eÉaggel©sasqa© me de± tŸn basile©an toÓ
qeoÓ). In Luke 5.1 and in three other verses in his gospel (8.11, 21; 11.28) Luke
reshapes his sources in order to emphasize that Jesus proclaimed ‘the word
of God’. In Acts this phrase frequently refers to post-Easter proclamation
and once (10.36) to Jesus’ own pre-Easter proclamation.
 Jesus and Gospel 61
The lines of continuity between pre- and post-Easter proclamation
drawn by all three synoptic evangelists are clear.145 Equally significant is
the fact that in differing ways the evangelists indicate that the gospel word
group is synonymous with ‘the word’ or ‘the word of God’. In section 2.7
I noted that in several passages in Paul’s earliest letter ‘the word’ (I Thess.
1.6, 8 and 2.13; cf. II Thess. 3.1) and ‘the gospel’ (I Thess. 1.5; 2.2, 4, 8, 9; 3.2;
II Thess. 1.8 and 2.14) are almost synonymous terms. At Gal. 1.23 Paul uses
‘the faith’ to refer to Christian proclamation (probably for the only time),
where ‘the gospel’ might well have been used. Similarly with Paul’s four
uses of t¼ kžrugma (‘proclamation’) (I Cor. 1.21; 2.4; 15.14; Rom. 16.25).
In short, as I emphasized at the outset, a basket of near-synonymous
words and phrases is used in early Christian writings to refer to God’s
good news or word concerning Jesus Christ. Only the gospel word group has
verbal links with the language of the imperial cult. Hence we must be wary of
assuming that with every early Christian use of this word group the imperial
cult is lurking in the background. The content of Christian proclamation,
however, was sufficiently close to the central themes of imperial propaganda
for it to be understood by many as a rival, and in due course as a threat.
In section 2.8 attention was drawn to the ways in which ‘the gospel’ was
used in earliest Christianity (and especially by Paul) as a shorthand term and
identity marker. The phrase was part of the social dialect of several strands
of early Christianity. Although Christian usage was quite distinctive, it was
not incomprehensible to ‘outsiders’. For it was of the very essence of the
whole semantic field associated with this noun that t¼ eÉagg”lion was
open proclamation, open heralding, to Jew and Gentile alike, of God’s
evangel or Gospel concerning Jesus Christ.
The pattern of usage I have sketched was replicated in a number of
other cases. Key early Christian words such as –kklhs©a, c†riv, ˆg†ph,
and katall†ssw were not unknown outside the walls of the earliest house
church communities, but they were developed considerably and linked with
other words and phrases in order to express the newness of the content of
the Gospel.

145 It is usually assumed that the Fourth Evangelist has merged without remainder proclamation of
Jesus and post-Easter proclamation, so that there is complete continuity between the two. This isn’t
quite the whole story, for the evangelist does distinguish between pre- and post-Easter by means
of his ‘memory’ motif. When Jesus says, ‘Destroy this temple, and in three days I will raise it up’
(2.19), neither the Jewish leaders nor the disciples understand what he means. However, the evangelist
explains that in the light of resurrection faith the disciples remembered this saying and understood
its significance (2.22). So too with the entry into Jerusalem (12.16; see also 14.26 and 20.9).
62 Jesus and Gospel
I have suggested that early Christian use of the noun ‘gospel’ was deve-
loped in a quite precise social setting. The relationship between the Chris-
tian Gospel and culture must always be two-sided. Language has to be found
which rings bells in any given cultural setting, but always with the recogni-
tion that Christian proclamation may be subversive, as it undoubtedly was
in the first century.
In my Inaugural Lecture in the University of London in 1978 I quoted
with approval Gerhard Ebeling’s hermeneutical dictum: ‘the same word can
be said to another time only by being said differently’. I want to repeat that
dictum today: the same Gospel can be said to another time only by being said
differently.146 So I do not want to suggest that Christians should necessarily
sprinkle the noun ‘gospel’ like confetti over all their mission statements,
for, in some strands of early Christianity, the noun is conspicuous by its
absence. My hope is that Christians will find new ways to express the concept
meaningfully. In taking up that challenge, we shall be helped by a keener
appreciation of the rich and diverse ways the noun ‘gospel’ is used in the
New Testament writings, and in particular by Paul.

146 G. Ebeling, ‘Time and Word’, in J. M. Robinson, ed., The Future of our Religious Past: Essays in
Honour of Rudolf Bultmann (E. tr. London: SCM, 1971), p. 265.
 chap t e r 3

The fourfold Gospel

The origins and the theological significance of the fourfold Gospel raise a set
of teasing questions.1 Why did the early church eventually accept four partly
parallel foundation documents, no more, no less? There is no precedent for
this either in the OT Scriptures or elsewhere in earliest Christianity. Did
retention of four gospels assist or hinder the early church in the presentation
of its claims concerning Jesus? No doubt, to some, insistence that there were
four gospels implied that there were basic flaws in the single gospels. Was
the second-century church’s decision to bring together four separate gospels
wise? What were, and what are, the theological implications of the fourfold
Gospel? A critical theology cannot avoid asking these questions.
In the early decades of the twentieth century, the views of the great giants,
Theodore Zahn and Adolf von Harnack, were influential: many scholars
accepted their view that the fourfold Gospel emerged very early in the se-
cond century, well before Marcion.2 More recently, particularly under the
influence of Hans von Campenhausen, most scholars have accepted that
the fourfold Gospel emerged in the second half of the second century
and that the Muratorian Fragment and Irenaeus are our primary witnesses.3
However, the current consensus on the emergence of the fourfold Gospel
is now being challenged from two entirely different starting points. The
1 This chapter was my Presidential Address delivered on 7 August 1996 at the 51st General Meeting
of Studiorum Novi Testamenti Societas in Strasbourg, France. It was published in NTS 43 (1997)
317–46, and has been lightly revised for this book.
2 A. von Harnack, The Origin of the New Testament and the Most Important Consequences of the New
Creation (London and New York, 1925), pp. 69–72; Th. Zahn, Grundriss der Geschichte des Neutesta-
mentlichen Kanons, 2nd edn (Leipzig: Deichert, 1904), pp. 35–41. E. J. Goodspeed dated the origin
of the fourfold Gospel to c. 125: see his The Formation of the New Testament (Chicago: University
of Chicago Press, 1937), pp. 33–41. Five years later John Knox rejected Goodspeed’s arguments and
opted for the West between 150 and 175 as the time when ‘we get our first glimpse of the existence of
the fourfold Gospel’, Marcion and the New Testament (Chicago: University of Chicago Press, 1942),
pp. 140–67. K. L. Carroll, ‘The Creation of the Fourfold Gospel’, BJRL 37 (1954–5) 68–77 echoed
Knox’s claim that the fourfold Gospel was an answer to Marcion.
3 Hans von Campenhausen, Die Entstehung der christlichen Bibel (Tübingen: Mohr, 1968); E. tr. The
Formation of the Christian Bible (Philadelphia: Fortress; London: Black, 1972), chapter 5.

63
64 Jesus and Gospel
Muratorian Fragment is being assigned by some to the fourth century,
and, as a corollary, Irenaeus’ devotion to the fourfold Gospel is seen as
‘something of an innovation’ in a time of fluidity of gospel traditions and a
proliferation of gospels.4 The other challenge to the consensus approaches
the question from a very different angle. Whereas the traditional way of
discussing this question focusses on the use early Christian writers made
of the four gospels, attention is now being given to the evidence of the
earliest copies of the gospels themselves – especially to the predilection of
Christian scribes for the codex and for nomina sacra.5
Theological reflection on the significance of Christianity’s commitment
to four gospels has been sparse in recent years. Very little has been written
since Oscar Cullmann’s important article first published in 1945.6 However,
in some circles a strong challenge has been mounted to the pre-eminence of
the canonical four in historical reconstructions of the origin and develop-
ment of early Christianity. Now and again this challenge is accompanied by
hints of a theological agenda: we are told that by giving increased attention
to non-canonical gospels it may be possible to construct a Jesus who will
be more congenial in a post-modernist era.
I shall attempt to take account both of the ways second-century writers
used, and referred to, the gospels and also of the evidence of the earliest
manuscripts. I shall work backwards from Irenaeus, for I find that it is often
helpful to work back from the full flowering of a concept or a development
to its earlier roots. I shall insist that the decision to accept four gospels,
along with the earlier acceptance of a plurality of gospels, was one of the
most momentous ones taken within early Christianity, a decision which
cries out for continuing theological reflection.

3.1 irenaeus
Book iii of Irenaeus’ Adversus Haereses was written about ad 180. Irenaeus
comments on the origin of the four individual gospels, which he clearly

4 See especially G. M. Hahneman, The Muratorian Fragment and the Development of the Canon (Oxford:
Clarendon, 1992), p. 101. Note also, ‘It is difficult therefore to acknowledge that the fourfold Gospel
was “firmly established” in the last quarter of the second century’ (p. 100; and cf. p. 108). Hahneman
develops considerably the arguments for a fourth-century date first advanced by A. C. Sundberg in
two articles: ‘Towards a Revised History of the New Testament Canon’, Studia Evangelica 4/1 (1968)
452–61; ‘Canon Muratori: A Fourth Century List’, HTR 66 (1973) 1–41.
5 See below, pp. 71–5.
6 O. Cullmann, ‘Die Pluralität der Evangelien als theologisches Problem im Altertum’, Theologische
Zeitschrift 1 (1945) 23–42; E. tr. in Cullmann’s The Early Church, ed. A. J. B. Higgins (London:
SCM, 1956), pp. 37–54. See also R. C. Morgan, ‘The Hermeneutical Significance of Four Gospels’,
Interpretation 33 (1979) 376–88; R. A. Burridge, Four Gospels, One Jesus? (London: SPCK, 1994),
pp. 163–79.
 The fourfold Gospel 65
accepts as ‘Scripture’, and sets out the earliest defence of the church’s four-
fold Gospel. His main point is clear: there is one Gospel in fourfold form,
held together by one Spirit (Adv. Haer. iii.11.8).
Irenaeus frequently refers to ‘the Gospel’, ‘the Gospel according to . . .’
and only very rarely to ‘four gospels’.7 The Gospel is primarily the faith
proclaimed and transmitted by the apostles, and only secondarily the writ-
ten record ‘reported’ by such and such an evangelist.8 If there is one Gospel,
why are there four written accounts of it? Why four, no more, no less?
Irenaeus’ attempt in iii.11.8 to defend the number ‘four’ with analogies
from both the natural and the spiritual worlds is well known. His fourfold
appeal to the four points of the compass and the four winds, the four-faced
cherubim of Ezekiel 1 and the four living creatures of Rev. 4.7, the fourfold
activity of the word of God, and God’s four covenants with mankind has
been derided as a ‘fundamental error’9 and seen as a quite desperate attempt
to defend a recent innovation.10
I do not think that this reading of Irenaeus is accurate: his discussion of
the fourfold Gospel is much more sophisticated than many writers have sup-
posed. All too often Irenaeus’ comments on the number four are wrenched
out of context; they are in fact a digression in a lengthy and often perceptive
discussion of the authority and reliability of the witness of the Scriptures
to one God, the Creator of all.
Irenaeus’ views on the four gospels are established long before he offers
reasons why there are four gospels, no more, no less, in chapter 11 of Book iii.
In the Preface to Book iii Irenaeus states clearly that the Gospel preached
by the apostles has been ‘handed on to us in the Scriptures, so that the
Gospel may be the foundation and pillar of our faith’. This image of
the Gospel as the ‘foundation and pillar’ of the church, an allusion to I
Tim. 3.15, is repeated in chapter 11 and extended to the four gospels as the
four pillars of the church.

7 See A. Benoit, Saint Irénée. Introduction à l’étude de sa théologie (Paris: Presses Universitaires de France,
1960). Benoit notes that, in Book iii, ‘gospel’ is used in the singular forty-one times; twelve times
for a particular gospel; only six times in the plural. See also Yves-Marie Blanchard, Aux sources du
canon, le témoinage d’Irénée, Cogitatio Fidei 174 (Paris: Cerf, 1993), p. 157, who counts seventy-five
occurrences of ‘gospel’ in Book iii, only five of which are in the plural.
8 See especially the Preface to Book iii, and iii.1.1. For a helpful discussion of the verbs used to refer to
the ‘reporting’ or ‘recording’ activity of the evangelists, see Blanchard, Aux sources du canon, p. 161.
9 Cullmann, The Early Church, pp. 50–2 claims that Irenaeus’ justification of the fourfold Gospel ‘is
based on the same fundamental error as the Gnostics’ “docetic” arguments against it’: his appeal
to four as a ‘divinely ordained number’ left out of account the purely human circumstances of the
formation of the fourfold Gospel.
10 Theodor Zahn, Geschichte des neutestamentlichen Kanons (2 vols., Erlangen: Deichert, 1888),
Vol. i, p. 153, scornfully rejects attempts to write off Irenaeus’ arguments as ‘dogmatic assertions
and theosophical trifles’.
66 Jesus and Gospel
In the important opening paragraphs of Book iii Irenaeus comments
further on the origins of the four gospels. After Pentecost the apostles
proclaimed the Gospel orally; two of the apostles and two of their followers
wrote gospels. Discussion of the human origins of the four written gospels
is followed by emphasis on their theological unity: ‘They have all declared
to us that there is one God, Creator of heaven and earth, announced by
the law and the prophets; and one Christ, the Son of God’ (iii.1.1-2).
So right from the outset of Book iii the reader knows that the church
has the one God-given Gospel as recorded by two apostles and two of their
immediate associates. In other words, the Gospel has been given to the
church in fourfold form, and chapter 11 with its set of four arguments,
within each of which the number four plays a central role, is hardly neces-
sary. We may even feel that the extended defence of the number four in 11.8
weakens rather than strengthens Irenaeus’ case, but his first readers proba-
bly thought otherwise, for they were accustomed to seeing hidden meaning
in numbers. At the outset of the Adversus Haereses Irenaeus summarizes the
Valentinians’ views and shows that the number four played an important
role in their speculations.
For Irenaeus’ readers, the number four would certainly have evoked
solidity and harmonious proportion, precisely his intention. As an example
of the evocative nature of the number four in Irenaeus’ day, let me mention
the Tetrapylon at Aphrodisias, completed just a few years before Irenaeus
wrote. This superb gateway to Aphrodisias, one of the finest and most
influential cities of the ancient world in the second century, has four recently
re-erected columns, each one of which has four richly decorated faces. So
Irenaeus’ first readers may well have been impressed by his claims that the
outward form of the Gospel should be harmoniously composed and well
proportioned, just like God’s creation (iii.11.9).
Now to a rather different point. Irenaeus is fascinated by the beginnings
of the four gospels. He refers to them three times in Book iii, beginning
at 11.9. Why does he cite and comment on the openings of the four gospels
so fully? He could have made his general point concerning one God the
Creator from many other passages from the gospels. The considerable vari-
ations in the openings of the four gospels must have baffled both Christian
and non-Christian alike. Irenaeus does state that the Valentinians seized on
the errors and contradictions of the gospels (iii.2.1).11 So, probably with an
11 ‘. . . in accusationem convertuntur ipsarum Scripturarum, quasi non recte habeant, neque sint ex
auctoritate et quia varie sint dictae . . .’. I have used the edition of the Latin text in Irénée de Lyon.
Contre les hérésies, iii, ed. A. Rousseau and L. Doutreleau, Sources Chrétiennes 211 (Paris: Cerf, 1974),
which includes the Greek fragments and a retroversion of the Latin into Greek.
 The fourfold Gospel 67
eye on his opponents, Irenaeus stresses that, in spite of their very different
starting points, the four gospels do have a theological unity.12 The Mura-
torian Fragment, to which we shall come in a moment, makes a similar
point.
Although Irenaeus often cites passages from the four gospels accurately,
he also regularly introduces sayings of Jesus with ‘the Lord said’, ‘the Lord
said in the Gospel’, ‘the Lord declared’, without indicating from which
particular gospel the sayings are taken. At the end of the Preface to Book iii,
for example, a version of Luke 10.18 is introduced with the words ‘the Lord
declared’. In this case, the text is cited in abbreviated form: it is difficult to
decide whether the variation occurs as the result of faulty memory, Irenaeus’
knowledge of an otherwise unattested textual tradition, or his use of oral
tradition. In the middle of his extended discussion of the opening chapters
of Luke’s Gospel, Irenaeus refers to four verses from John 1, but without
indicating that he has switched from Luke to John (iii.10.3). Matt. 12.18-21
is quoted as part of the discussion of the opening chapters of John’s Gospel,
but, once again, the reader is not told about the change of gospels. Similar
phenomena occur elsewhere. This is not surprising once we recognize that,
for Irenaeus, ‘the Gospel’ and in particular the words of Jesus have a higher
authority than the individual writings of the evangelists, even though the
gospels are referred to occasionally as ‘Scriptures’.
Irenaeus is able to cite the written gospels both carefully and carelessly,13
to weave together loosely passages from two or more gospels, and to intro-
duce sayings with ‘the Lord said’, some of which seem to be taken from
the written gospels, some from oral tradition. The fact that these various
phenomena are found in a writer for whom the fourfold Gospel is funda-
mental stands as a warning sign for all students of gospel traditions in the
second century. Earlier Christian writers may also value the written gospels
highly even though they appeal directly either to the words of Jesus or to
oral tradition, or even though they link topically sayings of Jesus taken from
two or more gospels. Irenaeus was not the only writer who cites ‘words of
the Lord’, and does not tell us whether he is quoting from written gospels
or from oral tradition.
By the time Irenaeus wrote in about ad 180, the fourfold Gospel was very
well established. Irenaeus is not defending an innovation, but explaining
why, unlike the heretics, the church has four gospels, no more, no less: she
12 See Zahn, Geschichte, Vol. ii, p. 43 and H. Merkel, Die Widersprüche zwischen den Evangelien. Ihre
polemische und apologetische Behandlung in der alten Kirche bis zu Augustin (Tübingen: Mohr, 1971),
pp. 42–3.
13 Matt. 11.27 is cited in three different ways at iv.6.1, 3, 7.
68 Jesus and Gospel
has received four written accounts of the one Gospel from the apostles and
their immediate followers.14

3.2 the muratorian fragment

The second pillar in most discussions of the origin of the fourfold Gospel
has been the Muratorian Fragment, generally dated to just before, or just
after, Irenaeus’ Adversus Haereses. Is this view still tenable, in spite of at-
tempts to date the Fragment to the fourth century? Or is this pillar begin-
ning to crumble?
In 1992 Geoffrey Hahneman picked up and developed a case for a fourth-
century date, first defended by A. Sundberg in 1968.15 Even though they
have now won a handful of converts, especially in the USA,16 I do not think
that the case for a fourth-century date has been made out.17 I shall do no
more than refer briefly to their three main points.
First, the cornerstone of the traditional dating has always been the Frag-
ment’s reference to Hermas’ composition of the Shepherd ‘nuperrime’, very
recently, in our times, in the city of Rome, while bishop Pius, his brother,
was occupying the episcopal chair of the church of the city of Rome (lines
73–6) – i.e. not all that long after ad 140.18 Hahneman, however, claims
that all the Fragment’s information about the Shepherd is ‘erroneous or
misleading’. He redates the Shepherd to c. ad 100 and then argues that
the Fragment’s attribution of the Shepherd to Hermas is fourth-century
pseudonymity designed to discredit the Shepherd. But, since the Shepherd

14 Cf. Benoit, Saint Irénée, p. 117: ‘La justification irénéenne ne veut pas être une démonstration,
elle ne fait qu’augmenter la crédibilité du fait accepté par ailleurs.’ T. C. Skeat, ‘Irenaeus and the
Four-Gospel Canon’, NovT 34 (1992) 193–9, claims that, in his celebrated identification of the four
evangelists with the four living creatures of the Apocalypse, Irenaeus has used an earlier source. His
case is strong but not conclusive, so I have not drawn on it in this chapter.
15 See n. 4 above.
16 In his article ‘Muratorian Fragment’ in The Anchor Bible Dictionary, ed. D. N. Freedman (New York:
Doubleday 1992), Vol. iv, p. 929, G. A. Robbins claims that Sundberg’s thesis ‘has won considerable
acceptance and further confirmation’. In his article ‘Canon, New Testament’ in the same dictionary,
H. Y. Gamble accepts Sundberg’s thesis cautiously (Vol. i, p. 856), as does H. Koester, Ancient Christian
Gospels: Their History and Development (London: SCM; Philadelphia: Trinity, 1990), p. 243.
17 See now J. Verheyden’s detailed study, ‘The Canon Muratori: A Matter of Dispute’, in J.-M. Auwers
and H. J. de Jonge, eds., The Biblical Canons (Leuven: Peeters, 2003), pp. 487–586. Verheyden
concludes (p. 556) that ‘the suggestion of a fourth century eastern origin for the Fragment should be
put to rest not for a thousand years, but for eternity’.
18 I have used the critical edition edited (with a facsimile reproduction) by S. P. Tregelles, Canon
Muratorianus (Oxford: Clarendon, 1867). Tregelles’s learned notes are still worth consulting. See
also H. Lietzmann’s edition, Das Muratorische Fragment und die monarchianischen Prologue zu den
Evangelien, Kleine Texte, i (Bonn, 1902).
 The fourfold Gospel 69
is still listed in the Fragment as ‘recommended reading’, this seems
implausible.19
Secondly, the Fragment is said to fit naturally into fourth-century
catalogues of canonical writings; it is an anomaly in the second century.
This line of argument is well off the mark, for the Fragment is not a canoni-
cal list or catalogue at all. Its genre is that of ‘Einleitung’ comments about
the origin and authority of early Christian writings; the only two later uses
of the Fragment are in prologues, not lists.20
Thirdly, the Fragment is allegedly out of line with other evidence for
the development of the canon. In my judgement none of the Fragment’s
comments is anomalous in a second-century setting; many fit much more
readily into that setting than into a fourth-century context.21 I shall now ex-
plore this point with reference to the Fragment’s comments on the gospels.
There is general agreement that the Fragment’s comments on Luke were
preceded by comments on Matthew and Mark. The second line could
well be a title for Luke’s Gospel: the third book of the Gospel according to
Luke.22 Just like Irenaeus, the Fragment uses both the formal phrase, ‘Gospel
according to Luke’, ‘evangelium secundum Lucam’, a direct translation of
eÉagg”lion kat‡ LoukŽn (a phrase to which I shall return later), and
also uses the plural, ‘fourth of the gospels’ (line 9, and similarly in lines 17
and 20).
The Fragment comments more fully on the origin of ‘the fourth of the
gospels’ than on any other writing.23 The attempt to link the origin of
this gospel to the whole apostolic circle smacks of apologetic: the Fourth
Gospel, it is claimed, stems ultimately from revelation. This thorough-
going defence of the Fourth Gospel would surely not have been needed in

19 Similarly, E. Ferguson in his critical review of Hahneman’s monograph in JTS 44 (1993) 691–7:
‘The pseudonymity would seem to be of doubtful value in a polemic against the Shepherd in the
fourth century.’ See also E. Ferguson’s discussion of Sundberg’s theory in ‘Canon Muratori: Date
and Provenance’, Studia Patristica 18 (1982) 677–83.
20 Cf. J.-D. Kaestli: ‘Par son contenu et par sa forme, le CM (Canon de Muratori) est plus proche du
genre des “prologues” que de celui des “listes canoniques”: ‘La Place du Fragment de Muratori dans
l’histoire du canon. A propos de la thèse de Sundberg et Hahneman’, Cristianesimo nella Storia 15
(1994) 609–34, here 616. Similarly Ferguson, JTS 44 (1993) 696.
21 For similar conclusions, see Ferguson, JTS 44 (1993) 691–7; Kaestli, ‘La Place du Fragment de
Muratori’; P. Henne, ‘La Datation du Canon de Muratori’, RB 100 (1993) 54–75; and W. Horbury,
‘The Wisdom of Solomon in the Muratorian Fragment’, JTS 45 (1994) 149–59.
22 A. T. Ehrhardt suggests that the comment about Luke in lines 6–7, ‘dominum tamen nec ipse vidit
in carne’, is sufficient evidence ‘to assume that Papias was responsible for the fragmentary remark
about St. Mark in the Muratorian Fragment’: ‘The Gospels in the Muratorian Fragment’, in his The
Framework of the New Testament Stories (Manchester: Manchester University Press, 1964), pp. 11–36,
here p. 13; this chapter was first published in German in Ostkirchen Studien 2 (1953) 121–38.
23 See especially Ehrhardt, Framework, pp. 18–25.
70 Jesus and Gospel
the fourth century, but we do know that in the latter part of the second
century there were doubts in some circles about the Fourth Gospel, most
notably among the Alogi and the followers of the anti-Montanist Gaius.
Andrew is the only apostle who is named at this point. This is not
surprising, since in John 1.40 Andrew is identified by name as the first
person to respond to John’s witness to Jesus. In the Fourth Gospel Andrew
is given a prominence which he does not have in the other three gospels.
As we shall see shortly, Papias also singles out Andrew for special mention
and in so doing reveals his knowledge of the Fourth Gospel.
The Fragment’s lengthy defence of the Fourth Gospel includes in lines
16–26 an important reference to the fourfold Gospel. The Fragment con-
cedes that different beginnings are taught in the various gospel books, yet
insists that they are held together by one primary Spirit. This is surely
a response to critics who have pounced on the different beginnings of the
gospels.24 As in other lines, we are close to Irenaeus, though there is no sign
of verbal dependence. I have already drawn attention to the way Irenaeus
comments at length on the beginnings of the gospels, probably partly in
response to critics. Similarly in the Fragment.25 In lines 7–8 the opening
of Luke’s story is referred to. In lines 16–26 a theological response is made
to the criticism that the gospels have different beginnings: by the one pri-
mary Spirit, the central themes of the story of Christ are found in all four
gospels. As with Irenaeus, the fourfold Gospel is not an innovation, but it
does need to be defended against the jibes of critics who poke fun at the
different openings of the gospels.
Who did this? As I noted above, Irenaeus refers to the Valentinians. I
also suspect that Celsus or some other pagan critic may well be lurking
behind Irenaeus’ comments and lines 16–26 of the Fragment. Writing be-
tween 177 and 180, just a few years before Irenaeus wrote Book iii of the
Adversus Haereses, Celsus knew all four gospels and had a particular interest
in their early chapters. According to Origen, Celsus’ Jew claimed that some
Christians, as if somewhat the worse from drink, ‘alter the original text of
the Gospel three or four or several times over, and they change its cha-
racter to enable them to deny difficulties in the face of criticism’.26 I take

24 So also R. M. Grant, The Earliest Lives of Jesus (London: SPCK, 1961), p. 31.
25 See also Zahn, Geschichte (above, n. 2), Vol. ii, p. 43.
26 Contra Celsum ii.27, ed. and trans. H. Chadwick (Cambridge: Cambridge University Press, 1953),
p. 90. See also v.56, where Origen responds to Celsus’ jibes concerning the number of angels at
the tomb of Jesus. For other evidence of the problems caused by differences in the gospels, see
H. Merkel, Die Widersprüche (above, n. 12), and Die Pluralität der Evangelien als theologisches und
exegetisches Problem in der alten Kirche (Bern: Peter Lang, 1978).
 The fourfold Gospel 71
this to be a reference to differences between the ‘three or four’ canonical
gospels.27
The Fragment refers to the two parousias of Christ and rather optimisti-
cally claims that this schema is found in all four gospels. ‘Everything is
declared in all the gospels . . . concerning his two comings, the first in hu-
mility when he was despised, which is past, the second, glorious in royal
power, which is still in the future.’ This schema is first developed fully by
Justin Martyr, though I have argued that it is partly anticipated in Matthew’s
Gospel.28 The two parousias schema is very prominent in Justin’s writings;
it is also found in his Apology, in Irenaeus, Tertullian, Hippolytus, Origen,
and the Anabathmoi Iakobou, but not, as far as I can discover, in fourth-
century writings.
The Fragment confirms that the fourfold Gospel was well established
towards the end of the second century.29 Quite independently, the Fragment
and Irenaeus make similar points concerning the fourfold Gospel: in spite
of what critics may say about the different beginnings of the gospels, there
is one Gospel in fourfold form, held together by one Spirit.30 Needless to
say, the points I have emphasized are conspicuous by their absence in the
recent attempts to locate the Fragment in the fourth century.

3.3 early four-gospel codices

Could Irenaeus and the writer of the Muratorian Fragment have used
codices containing all four gospels? In 1968 von Campenhausen denied
vigorously that the references to the fourfold Gospel in Irenaeus and
the Muratorian Fragment had anything to do with ‘book production’ or
Christian use of a four-gospel codex.31 In 1933, however, F. G. Kenyon had
edited the recently discovered ∏45, the Chester Beatty codex of the four
gospels and Acts, and had noted that this new evidence meant that it was
27 So too Merkel, Die Widersprüche, p. 11. See also T. Baarda, ‘DIAFWNIA – SUMFWNIA: Factors in
the Harmonization of the Gospels, Especially in the Diatessaron of Tatian’, in W. L. Petersen, ed.,
Gospel Traditions in the Second Century (Notre Dame and London: University of Notre Dame Press,
1989), pp. 133–5, reprinted in his Essays on the Diatessaron (Kampen: Kok Pharos, 1994), pp. 29–48;
Grant, Earliest Lives, pp. 59–60.
28 G. N. Stanton, ‘The Two Parousias of Christ: Justin Martyr and Matthew’, in M. C. de Boer, ed.,
From Jesus to John, FS M. de Jonge (Sheffield: JSOT Press, 1993), pp. 183–96.
29 Ehrhardt, Framework, p. 11 suggests that the Fragment was produced at Rome, probably under
Zephyrinus, ad 197–217.
30 I am not convinced by A. Ehrhardt’s claim (ibid., pp. 14–15) that the reference to John as ‘ex discipulis’
in line 16 betrays its origin in Irenaeus, who refers to John the evangelist as ‘the disciple of the Lord’,
but never refers to him as the son of Zebedee; this similarity could have arisen from independent
use of the same tradition.
31 Von Campenhausen, Formation (above, n. 3), pp. 173–4.
72 Jesus and Gospel
possible to believe that Irenaeus may have been accustomed to the sight of
codices which contained all four gospels.32 Kenyon dated the codex to the
first half of the third century and concluded that it was the earliest example
of a codex containing all four gospels.
Kenyon’s dating of ∏45 and his general observations on this codex have
been widely accepted, though most scholars have overlooked his impor-
tant comment on Irenaeus’ probable knowledge of four-gospel codices. In
recent years our knowedge of Biblical papyri and codices has increased very
considerably: we now have fairly solid evidence for two further four-gospel
codices which are even earlier than ∏45.
∏75, the Bodmer papyrus of Luke and John, has attracted plenty of
attention from text critics. There is general agreement that it dates from
early in the third century; the editors dated it to between ad 175 and
225. However, the possibility that its fragments of Luke and John formed
the second of two single-quire codices sewn together does not seem to
have been considered until 1994.33 Why would Luke and John be bound
together without Matthew and Mark? It is just possible to imagine a codex
containing Matthew and John, the two gospels considered to have been
written by apostles. But a codex containing only Luke and John is most
unexpected. In fact we have no other example of a two-gospel codex.34
T. C. Skeat has calculated that ∏75 contained seventy-two leaves, and
noted that a codex of double this size would have been almost impossible to
handle. So he concludes that ∏75 may have originally consisted of a single-
quire codex containing Matthew and Mark, sewn together with another
containing Luke and John, and then bound.35
A strong case has recently been made out by T. C. Skeat for an even ear-
lier four-gospel codex.36 ∏64, the fragments of Matthew held at Magdalen
College, Oxford, was first edited by Colin Roberts in 1953 and dated to
32 The Chester Beatty Biblical Papyri, Fasciculi i and ii, The Gospels and Acts, General Introduction and
Text (London: Emery Walker, 1933), here i, p. 13. ∏45 is made up of quires of two leaves – a single
sheet of papyrus folded into two. The letters are small; the scribe was no calligrapher.
33 T. C. Skeat, ‘The Origin of the Christian Codex’, ZPE 102 (1994) 263–8, here 264.
34 ∏53, third century, with fragments of Matthew and Acts in the same hand, is an interesting possible
partial exception, though the fragments may not be from the same codex. If they are, and if the
codex included all four gospels, it would have run to a highly improbable 300–50 leaves. So a codex
with one or two gospels, plus Acts, is possible. Cf. K. Aland, Repertorium der griechischen christlichen
Papyri, Vol. i (Berlin and New York: De Gruyter, 1976), pp. 53 and 283. See also J. van Haelst,
Catalogue des papyrus littéraires juifs et chrétiens, Papyrologie i (Paris: Publications de la Sorbonne,
1976), nos. 380, 381.
35 Skeat, ‘Origin’, p. 264. In ‘The Oldest Manuscript of the Four Gospels?’ NTS 43 (1997) 31, Skeat
mentions in passing the possibility that perhaps there were two volumes, i.e. two single-quire codices
bound separately.
36 Skeat, ‘Oldest Manuscript’, pp. 1–34.
 The fourfold Gospel 73
the late second century on the grounds that its hand is a precursor of the
style known as Biblical Uncial or Biblical Majuscule. These are the three
fragments which have attracted so much attention in the media since a sen-
sationalist article published in The Times of London on 24 December 1994
reported C. P. Thiede’s claim that they date from the mid first century.37
There has never been any doubt that ∏67, with its further fragments of
Matthew now in Barcelona, was from the same codex as ∏64. For some
six months before the media became interested in ∏64, T. C. Skeat had
been working intensively on ∏64 + ∏67, as well as ∏4, fragments of
Luke in the Bibliothèque nationale in Paris. Skeat has now shown beyond
reasonable doubt that ∏64 + ∏67 + ∏4 are from the same single-quire
codex, probably our earliest four-gospel codex, which may date from the
late second century.
In his NTS article Skeat does not comment on the significance of one of
the codex’s most striking features, its double columns.38 The format of two
columns to the page is rare in papyrus codices. This is the only example of a
two-column Greek New Testament papyrus manuscript, though there are
four examples in early fragments of OT papyri.39 The narrow columns, with
only about fifteen letters in each column, would have assisted reading aloud
in the context of worship.40 The columns have about the same number of
letters as each of the three columns of Codex Vaticanus, and only about

37 See G. N. Stanton, Gospel Truth? New Light on Jesus and the Gospels (London: HarperCollins,
1995), pp. 1–19. For Thiede’s theory, see his ‘Papyrus Magdalen Greek 17 (Gregory–Aland ∏64): A
Reappraisal’, ZPE 105 (1995) 13–20; reprinted in TB 46 (1995) 29–42. The theory has been rejected by
numerous scholars, all of whom accept Roberts’s original date for ∏64, the late second century. See
especially J. Neville Birdsall, ‘The Dating of the Magdalen Papyrus’, Church Times, 6 January 1995;
Klaus Wachtel, ‘64/67: Fragmente des Matthäusevangeliums aus dem 1. Jahrhundert?’, ZPE 107
(1995) 73–80; Peter M. Head, ‘The Date of the Magdalen Papyrus of Matthew (P. Magd. Gr. 17 =
P64): A Response to C. P. Thiede’, TB 46 (1995) 251–85; D. C. Parker, ‘Was Matthew Written before
50 CE? The Magdalen Papyrus of Matthew’, Expository Times 107 (1995) 40–3; S. R. Pickering in NT
Textual Research Update 2 (1994) 94–8 and 3 (1995) 22–5; P. Grelot, ‘Remarques sur un manuscript
de l’Evangile de Matthieu’, RSR 83 (1995) 403–5; J. K. Elliott in NovT 38 (1996) 393–9; H. Vocke,
‘Papyrus Magdalen 17 – Weitere Argumente gegen die Frühdatierung des angeblichen Jesus-Papyrus’,
ZPE 113 (1996) 153–7.
38 After I had completed the research summarized in this paragraph, T. C. Skeat wrote to me (10 July
1996) as follows: ‘The two-column format became the standard form throughout the whole of the
Middle Ages, and has survived almost down to the present day in printed bibles and prayer-books.
Why? I have never seen this considered, but I suppose the answer must be that this is the easiest to
read. Certainly reading the lessons from ∏45 with its very long lines and small script would have
been quite difficult, and no doubt it wasn’t intended for liturgical use.’
39 See E. G. Turner’s list of papyrus codices written in two columns, The Typology of the Early Codex
(Philadelphia: University of Pennsylvania Press, 1977), p. 36.
40 In 1970 the NEB was published with one single wide column on each page. In 1989 the REB reverted
to the more traditional two-column format, partly in order to facilitate reading aloud in the context
of worship.
74 Jesus and Gospel
three fewer than the four columns of Codex Sinaiticus. The two great
fourth-century manuscripts were clearly intended for liturgical use. So
the use of two columns in ∏64 + ∏67 + ∏4 is almost certainly an
indication of a high-class codex, a splendid ‘pulpit edition’ intended for
liturgical use.41
There are several other indications that this codex was an édition de luxe.42
The codex was planned and executed meticulously: the skill of the scribe
in constructing it is most impressive.43 All these features indicate a most
handsome edition of the four gospels, which would have been expensive
to produce. This codex does not look at all like an experiment by a scribe
working out ways to include four gospels in one codex: it certainly had
predecessors much earlier in the second century.
The three early codices are not clones of one another, for they are all
constructed and executed quite differently.44 In all probability they had
a number of predecessors. So well before the end of the second century
there was a very well-established tradition of four-gospel codices. All three
papyrus codices were found in Egypt; the evidence of Irenaeus and the
Muratorian Fragment points towards the West. So the fourfold Gospel
seems to have been well established in both East and West at the end of the
second century, and probably very much earlier.
However, for two reasons some caution is necessary. First, it is just possi-
ble that one or more of these three codices was in fact written in the West.
If that seems an unlikely scenario, we need to bear in mind that a fragment
of Irenaeus, P. Oxy. 405 (from a roll) travelled from Lyons to Oxyrhynchus
within twenty years of its production, ‘not long after the ink was dry on
the author’s manuscript’, to quote Roberts’s memorable comment.45
Secondly, papyrologists are always, rightly, extremely cautious about dat-
ing handwriting styles and developments in the production of manuscripts.
On the other hand, the recent media attention has ensured that the dating

41 Cf. Turner, Typology, pp. 36–7. Van Haelst, Catalogue, no. 336, even says that this is the oldest
example of a codex with two columns; presumably he means the oldest Biblical codex, though, in
the light of Turner’s list, this is a doubtful claim.
42 This is T. C. Skeat’s phrase, ‘Oldest Manuscript’, p. 26.
43 C. H. Roberts noted that ‘in its handsome script as well as in its organization . . . it is a thoroughgoing
literary production’. Manuscript, Society and Belief in Early Christian Egypt (London: Oxford Univer-
sity Press for The British Academy, 1979), p. 23. Although Skeat, ‘Oldest Manuscript’, 2 and 7, has
been unable to find any trace of two of the examples given by Roberts – three different positions for
punctuation as well as omission and quotation signs – he concurs with Roberts’s general conclusions.
44 ∏45, the latest of the three, is the least impressive hand. Whereas ∏45 is made up of quires of two
leaves – a single sheet of papyrus folded in two – ∏75 probably contained two single-quire codices
bound together. ∏64 + ∏67 + ∏4 is a two-column, single-quire codex.
45 Roberts, Manuscript, p. 53; Roberts notes some similarities with ∏64 + ∏67 + ∏4, p. 23.
 The fourfold Gospel 75
of ∏64 + ∏67 + ∏4 has been considered carefully by several papyrologists.
Working independently, they have all dated this codex to the end of the
second century.46

3.4 earlier roots

Irenaeus, the Muratorian Fragment, and the three earliest codices of
the gospels all suggest that, by the latter decades of the second century
and probably much earlier, the fourfold Gospel was well established and
widely accepted. At what point, and under what circumstances, were four
gospels brought together? Since this was a major development within early
Christianity, our critical instincts encourage us to search for answers. Before
we resume the search, however, it is important to recall that there are no
explicit comments on the fourfold Gospel before Irenaeus, and to note that
knowledge and use of a plurality of gospels is not necessarily to be equated
with acceptance of the fourfold Gospel.
Is it possible to trace the roots of the development of the fourfold Gospel
to the first half of the second century? In discussion of this question, three
issues are particularly important.

(1) I turn first to Justin Martyr, whose knowledge and use of the gospels
shortly after the middle of the century is still much disputed, in spite of
intense research and debate. I am concerned with only one issue: did Justin
anticipate the adoption of the fourfold Gospel, or did he anticipate his
pupil Tatian’s harmony?
In Justin’s well-known account of eucharistic worship in I Apol. 67, he
refers to the reading of ‘the memoirs of the apostles or the writings of the
prophets, as long as time allows’. Here apostolic writings are being accorded
a similar authority to that of the writings of the prophets, which, as for
Irenaeus, is a Christian short-hand way of referring to the OT Scriptures.
But what are the ‘memoirs of the apostles’? In the preceding chapter the
reader is told explicitly that they are ‘the gospels’ (I Apol. 66), the first
Christian occurrence of the plural. This is not a later gloss, for, as Luise
Abramowski has shown, Justin does add similar explanatory phrases for his
readers.47
In addition to these two references to ‘the memoirs of the apostles’ in
the Apology, Justin uses the phrase thirteen times in one section of the
46 See above, n. 37.
47 L. Abramowski, ‘Die “Erinnerungen der Apostel” bei Justin’, in P. Stuhlmacher, ed., Das Evangelium
und die Evangelien (Tübingen: Mohr, 1983), pp. 341–54, esp. p. 341.
76 Jesus and Gospel
Dialogue, chapters 98–107, in which he seems to have incorporated his
own, earlier extended anti-gnostic exposition of Psalm 22, and in which he
emphasizes the ‘writtenness’ of ‘the memoirs of the apostles’. At one point
in this exposition Justin refers to Peter’s memoirs; from the context this is
a reference to Mark’s Gospel (Dialogue 106.3). So in both the Apology and
the Dialogue the ‘memoirs’ are identified as written gospels.
How many gospels does Justin accept? In Dialogue 103.8 he refers to
‘the memoirs composed by his apostles and those who followed them’ (–n
g‡r to±v ˆpomnhmoneÅmasin, Œ jhmi Ëp¼ tän ˆpost»lwn aÉtoÓ kaª
tän –ke©noiv parakolouqhs†ntwn suntet†cqai . . .). Although Justin
never refers to the number of the gospels he accepts, this passage implies
that there were at least four. It is surprising how many recent writers have
ignored this point.48 There is general agreement that Justin used Matthew
and Luke regularly, and that Mark’s Gospel is referred to once (Dialogue
106.3). Justin’s knowledge of the Fourth Gospel is much disputed, but I am
convinced that I Apology 61.4 draws on John 3.3-5, and that Dialogue 88.7
shows knowledge of John 1.19-20. Justin’s failure to refer to John’s Gospel
more frequently is puzzling, but it may be related to his strong interest
in infancy narratives, and in ethical teaching and futurist eschatological
sayings – all in somewhat short supply in this gospel. Since there is no clear
evidence for Justin’s knowledge of any gospels other than the canonical
four, we can be all but certain that he had in mind Matthew, Mark, Luke,
and John, no more, no less.
Justin uses the singular ‘Gospel’ in only two passages, but in both cases
he is referring to written traditions. At Dialogue 10.2 Justin’s opponent
Trypho states that he has read with appreciation the commands of Jesus ‘in
the so-called Gospel’. At Dialogue 100.1 there is a similar usage: a citation of
Matt. 11.27 is introduced with the words, ‘in the Gospel it is written . . .’
(–n tä€ eÉaggel©w„ g”graptai e«pän . . . ) These two references recall
Irenaeus’ much more frequent use of the phrase ‘in the Gospel’. For Justin,
as for Irenaeus, the sayings of Jesus are of special importance: they are
recorded ‘in the Gospel’, ‘in the memoirs of the apostles’.
Unlike Irenaeus, Justin is not interested in the authorship or distinctive
features of the individual gospels. However, like Irenaeus, Justin knows at
least four written ‘memoirs’ or gospels, which can be referred to collectively
as ‘the Gospel’. Of course, Justin does not have Irenaeus’ clear concep-
tion of the fourfold Gospel, but the references in his extant writings to

48 It was noted already by S. P. Tregelles in 1867 (Canon Muratorianus (above, n. 18), p. 71): ‘no smaller
number [than four] could be implied by the two groups’.
 The fourfold Gospel 77
written gospels suggest that he may well have had a four-gospel codex in
his catechetical school in Rome by about ad 150.49
At this point account must be taken of the ways Justin cites sayings of
Jesus. The textual evidence is undeniably complex, and it is not easy to
account for the variations in wording from Matthew and Luke.50 In his
important recent study W. L. Petersen has shown that some of Justin’s
harmonized traditions can be traced in his pupil Tatian’s more thorough-
going harmony.51 Helmut Koester has gone further and suggested that
‘Justin was composing the one inclusive new Gospel which would make its
predecessors, Matthew and Luke (and possibly Mark), obsolete’.52 If one
focusses attention on the wording of the citations, Justin’s use of, or even
composition of, a harmony of sayings of Jesus is undeniable.
But how can such a conclusion be squared with Justin’s references to writ-
ten gospels? I think it is likely that for catechetical purposes (and possibly
even to disarm critics)53 Justin himself gathered together topically harmo-
nized clusters of sayings of Jesus from written gospels, primarily Matthew
and Luke.54 In this respect he partially anticipates Tatian, but I do not
believe that the corollary is an intention to do away with the ‘memoirs
of the apostles’, i.e. the written gospels in which the Saviour’s words were
recorded (cf. Dialogue 8.2).
The opening programmatic exchange between Justin and Trypho in the
Dialogue strongly suggests that Justin’s references to the sayings of Jesus
are based on written gospels. Justin recalls his own conversion experience
and the passionate desire which possessed him for the prophets, and for
those great men who are ‘the friends of Christ’, surely the apostles (8.2).
Justin refers immediately to the ‘dreadful majesty’ of the Saviour’s words
and the importance of carrying them out: there is a clear implication that
the words of Jesus have been written down by ‘the friends of Christ’. Trypho
responds with a taunt: Justin has been deceived, for he has been following

49 E. J. Goodspeed (Formation, p. 38) is much less cautious: ‘Justin became a Christian at Ephesus as
early as 135 ad, and probably there became attached to the fourfold gospel.’
50 See especially, A. J. Bellinzoni, The Sayings of Jesus in the Writings of Justin Martyr (Leiden: Brill,
1967) and Koester, Ancient Christian Gospels, pp. 360–402.
51 See W. L. Petersen, ‘Textual Evidence of Tatian’s Dependence upon Justin’s APOMNH-
MONEYMATA’, NTS 36 (1990) 512–34.
52 H. Koester, ‘The Text of the Synoptic Gospels in the Second Century’, in W. L. Petersen, ed., Gospel
Traditions in the Second Century (Notre Dame and London: University of Notre Dame Press, 1989),
p. 30, and cf. p. 32.
53 W. L. Petersen mentioned the latter possibility to me in a letter dated 1 September 1996.
54 Similarly Bellinzoni, Sayings of Jesus; E. Osborn, Justin Martyr (Tübingen: Mohr, 1973), p. 132; and
L. Abramowski, ‘Die “Erinnerungen” ’, 352–3.
78 Jesus and Gospel
men (plural) of no account, once again, surely, a reference to the apostles
who have recorded the words of the Saviour.
Like Irenaeus, Justin set great store by the words of Jesus. So for cate-
chetical purposes he seems to have used written gospels to make his own
harmonized collections of sayings of Jesus, linking them together topically.
On the other hand, Justin’s knowledge and use of four written gospels is
clear. Although in some respects he anticipates Tatian, in the use of written
gospels alongside harmonized sayings of Jesus his successor is Irenaeus. In
the light of our earlier conclusions, Justin’s reference to at least four writ-
ten gospels in Dialogue 103.8 suggests that he may well have possessed a
four-gospel codex in the library of his catechetical school.
(2) Martin Hengel has drawn attention to Zahn’s and Harnack’s views
on the titles of the gospels.55 In my judgement all three scholars have
correctly insisted that from early in the second century there was a profound
conviction that there was one Gospel ‘according to’ individual evangelists.
The evidence is so strong and so widespread that here we are surely in touch
with another of the roots of Irenaeus’ conviction that there is one Gospel
in fourfold form.
Hengel rightly attached weight to the evidence of the papyri. The open-
ing and closing leaves of papyri codices are usually missing, so the clear
inscriptio eÉagg”lion kat‡ ’Iw†nnhn (‘the Gospel according to John’) in
∏66 from about 200 is striking. It is in the same hand as the rest of the text,
but it has been added to the opening page a little awkwardly; the subscrip-
tio would have been identical. In ∏75, perhaps only a couple of decades
later, we have two examples on the same page of eÉagg”lion kat† . . . , a
subscriptio to Luke and an inscriptio to John.56
The evidence of ∏66 and ∏75 is consistent with the evidence of Irenaeus
and the Muratorian Fragment: in the second half of the second century
in many circles there was a strong conviction that there was one Gospel,
according to a particular evangelist. But what about the first half of the
second century? Helmut Koester rejects Martin Hengel’s theory that from
early in the second century the gospels must have had eÉagg”lion kat† . . .
attached to them as titles by claiming that Hengel has anachronistically read

55 M. Hengel, ‘The Titles of the Gospels and the Gospel of Mark’, in Studies in the Gospel of Mark
(London: SCM, 1985), pp. 64–84.
56 M . Hengel also notes that a page from ∏64 + ∏67 + ∏4, ‘which belong together’, has the inscriptio
eÉagg”lion kat‡ Maqqa±on (‘Titles’, p. 66). However, this example should not be set alongside
the evidence of ∏66 and ∏75, for the inscriptio (which is now located with the ∏4 fragments of
Luke) is not in the same hand; it probably comes from a fly-leaf added to the codex at a later point.
 The fourfold Gospel 79
back to the beginning of the century evidence of papyri from the end of
the second century.57
However, Koester and Hengel agree on one crucial point: as soon as
Christian communities regularly used more than one written account of
the actions and teaching of Jesus, it would have been necessary to distinguish
them by some form of title, especially in the context of readings at worship.58
That first happened as soon as Matthew had completed his writing, for
many Christians then had two accounts of the story of Jesus, Matthew’s
and Mark’s.
There is plenty of evidence for use of a plurality of gospels in many circles
in the first half of the second century. Papias, now dated to c. 110 by several
scholars, will serve as an example. He certainly knew Matthew and Mark.
I am convinced that Papias also knew John: there is no other reasonable
explanation for his list of disciples in the order, Andrew, Peter, Philip, and
Thomas – precisely the order in which they appear in the Fourth Gospel,
an order found nowhere else, though Andrew is singled out in line 14 of
the Muratorian Fragment as the apostle who received the revelation that
John, ‘ex discipulis’, should write down ‘quartum evangeliorum’.
So when early Christian communities used more than one gospel, how
were they differentiated, particularly in the context of worship? What are
the possible terms which could have been used to distinguish what we now
know as Matthew from Mark? Certainly not b©ov (‘life’), for which there is
no evidence; and not Justin’s ˆpomnhmoneÅmata (‘memoirs’), which was
not used by any Christian prior to Justin.59
As far as I can see, there is only one candidate, eÉagg”lion. As is well
known, in the first half of the second century it is not always easy to decide
whether eÉagg”lion refers to oral proclamation or to a written account
of the actions and teaching of Jesus. However, we may be confident that
eÉagg”lion refers to a writing in four passages in the Didache (8.2; 11.3;
twice in 15.3-4) which cannot be written off as late second-century redaction;
twice in Ignatius’ letter to the Smyrnaeans (5.1 and 7.2), and also in II
Clement 8.5.60 Once eÉagg”lion began to be used for a writing, it was a
natural extension to use this term as a title. What more appropriate way

57 H. Koester, ‘From the Kerygma-Gospel to Written Gospels’, NTS 35 (1989) 361–81, here 373 n. 2.
58 Cf. H. Koester, ‘Kerygma-Gospel’, p. 381 n. 1 and Hengel, ‘Titles’, pp. 74–81.
59 See Grant, Earliest Lives (above, n. 24), pp. 119–20.
60 For fuller discussion, see above pp. 54–5, and G. N. Stanton, ‘Matthew: BIBLOS, EUAGGELION,
or BIOS?’, in F. Van Segbroeck et al., eds., The Four Gospels 1992, FS Franz Neirynck, Vol. ii
(Leuven: Leuven University Press, 1992), pp. 1187–1202. For a different view, see R. H. Gundry,
‘EUAGGELION: How Soon a Book?’, JBL 115 (1996) 321–5.
80 Jesus and Gospel
was there of referring to individual gospels than as eÉagg”lion kat† . . .?
So one of the roots of the fourfold Gospel was undoubtedly the very
early use of the term eÉagg”lion to refer to a written gospel, and the
strong conviction that there was one Gospel, ‘according to’ a particular
evangelist.
(3) The very early separation of Luke and Acts is another indication of the
deep roots of the fourfold Gospel.61 It is generally accepted that Luke and
Acts were originally written on two separate rolls: they could not have been
squeezed onto one roll; the short Preface to Acts with its re-dedication to
Theophilus was a conventional way of introducing the second roll of a
single work. Once Christian scribes began to use the codex early in the
second century, it would have been possible for Luke and Acts to have
been juxtaposed in the same codex, with or without other writings, but,
as far as we know, this never happened.62 Luke and Acts are separated in
the Muratorian Fragment, and in all the lists and catalogues of canonical
writings. Irenaeus is the first writer to stress the close relationship between
Luke and Acts: he insisted that, if his opponents accepted Luke’s Gospel,
they should also accept Acts (Adv. Haer. iii.14.3 and 4).
We are so accustomed to treating Luke and Acts as one single writing
in two parts that it is easy to overlook the fact that in the second century
Luke’s Gospel and Acts circulated separately. Even in later centuries they
were not brought together. Two explanations are currently given for the
early separation of Luke and Acts. W. A. Strange has recently proposed that,
at Luke’s death, Acts remained in draft form; it remained in obscurity until
published in the third quarter of the second century, following editorial
work by both a ‘western’ and a ‘non-western’ editor.63 Even if this is a
plausible solution to the textual problems of Acts, I do not think that
Acts was unknown until the time of Irenaeus. Neither Irenaeus nor the
Muratorian Fragment presses a case for accepting Acts; they both imply that
the existence and authority of Acts had long been recognized. Irenaeus’ use
of Acts in polemic against his opponents would have been self-defeating
if Acts had only recently become available, for Irenaeus insists that, unlike

61 For discussion of the relationship of Luke and Acts, see M. C. Parsons and R. I. Pervo, Rethinking
the Unity of Luke and Acts (Minneapolis: Augsburg Fortress, 1993).
62 ∏53 is a possible exception; see n. 32 above. B. M. Metzger, The Canon of the New Testament: Its
Origin, Development, and Significance (Oxford: Clarendon, 1987), p. 296, lists a few late examples of
Luke coming fourth in a sequence of the four gospels, perhaps from a desire to bring the two books
by Luke side by side.
63 W. A. Strange, The Problem of the Text of Acts (Cambridge: Cambridge University Press, 1992), esp.
pp. 181–9.
 The fourfold Gospel 81
some of the writings used by heretics, the writings the church accepts are
ancient.64
I think that an alternative explanation is much more likely: the acceptance
of Luke into the fourfold Gospel led to its early separation from Acts,
probably before Marcion.65 There was plenty of interest in the second
century in the apostles, but even more interest in the sayings and actions
of Jesus recorded ‘in the Gospel’. Hence Acts seems to have remained
somewhat in the shadow of ‘the Gospel’.

Taken cumulatively, this evidence suggests that the adoption of the fourfold
Gospel may well have taken place in some circles (though not necessarily
everywhere) shortly before Justin’s day. Before I comment further on the
date of the emergence of the fourfold Gospel, I shall refer to explanations
which have been advanced for this momentous development within early
Christianity.

3.5 how and when?

What were the key factors which led to emergence of the fourfold Gospel?
Which form of explanation is preferable, a ‘big bang’ theory, or a theory of
gradual development?
It has often been urged that the fourfold Gospel was adopted in order to
counter the rapid growth and success of various groups of heretics, especially
gnostics. Of course, the production and use of gospels by gnostics may have
encouraged ‘the great church’ to clarify its position. But if heretics were
primarily in view, would it not have been wiser to opt for just one gospel?
Why four? When Irenaeus attacked the Valentinians of his day, he had to
show that all four gospels supported the theological point he was making:
it is hard to see how four gospels gave him a stronger case than one.
The same point is relevant when assessing whether or not the fourfold
Gospel was an answer to Marcion.66 It is hard to see how four gospels
assisted the struggle against Marcion. Von Campenhausen’s judgement is
surely correct: ‘That the new Gospel canon was particularly directed against
Marcion cannot be deduced from its composition.’67
64 See Benoit, Saint Irénée (above, n. 7), pp. 122–7 for a fuller discussion.
65 Similarly, W. R. Farmer and D. M. Farkasfalvy, The Formation of the New Testament Canon (New
York, 1983), p. 73, though they date the emergence of the fourfold Gospel somewhat later.
66 See Knox, Marcion (above, n. 2); so also Carroll, ‘Creation’ (above, n. 2).
67 Von Campenhausen, Formation, p. 171 n. 113. Elsewhere von Campenhausen accepts that Marcion
was very influential on the emergence of the NT canon; for discussion, see F. Bovon, ‘La Structure
canonique de l’Evangile et de l’Apôtre’, Cristianesimo nella Storia 15 (1994) 559–76.
82 Jesus and Gospel
A political explanation has often been advanced. The fourfold Gospel
is said to have been a compromise worked out between different regional
preferences.68 However, it is difficult to find any evidence for the second-
century equivalent of the European parliament! The older view that in-
dividual gospels circulated only in limited geographical areas is no longer
tenable:69 the papyri, both Christian and non-Christian, indicate clearly
that there was a great deal of contact between different regions around the
Mediterranean.70
A very different view is preferable. I am convinced that the emergence
of the fourfold Gospel is related to Christian adoption of the codex, for
no roll could contain four gospels. This explanation is hardly new: in 1933
F. G. Kenyon noted that the Chester Beatty papyri confirmed that the
Christian community was addicted to the codex rather than to the roll, and
recognized that, when the four gospels were brought into one codex, they
were ‘marked off as a single unit’.71 But Kenyon’s important observation
went largely ignored until C. H. Roberts and T. C. Skeat set out in full the
evidence for early Christian use of the codex and showed that Christian
scribes borrowed a Roman invention which had not been an immediate
success.72
The statistics are astonishing: among non-Christian papyri, rolls pre-
dominate until early in the fourth century, but Christian papyri are very
nearly all fragments of codices. With only one possible exception, every
single papyrus copy of the gospels is from a codex.73

68 Farmer and Farkasfalvy, Formation, emphasize the importance of the decision taken by Irenaeus
and Anicetus to agree to disagree over the date of Easter: ‘there was no other moment in Church
history when it is more likely that the fourfold Gospel canon was, in principle, implicitly agreed
upon’, p. 72. D. Trobisch also emphasizes the importance of this decision, reported by Eusebius, HE
v.24.14: Die Endredaktion des Neuen Testaments. Eine Untersuchung zur Entstehung der christlichen
Bibel (Freiburg: Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 1996), pp. 158–9.
69 So too von Campenhausen, Formation, p. 123: ‘It is highly questionable whether the idea is correct,
that originally each individual Gospel had its own territorial domain.’
70 See especially E. J. Epp, ‘New Testament Papyrus Manuscripts and Letter Carrying in Greco-Roman
Times’, in B. A. Pearson, ed., The Future of Early Christianity, FS H. Koester (Minneapolis: Fortress,
1991), pp. 35–56; E. J. Epp, ‘The Papyrus Manuscripts of the New Testament’, in B. D. Ehrman and
M. W. Holmes, eds., The Text of the New Testament in Contemporary Research, FS B. M. Metzger
(Grand Rapids: Eerdmans, 1995), pp. 3–21, here pp. 8–10.
71 Kenyon, Chester Beatty, Fasciculus i, pp. 12f.
72 C. H. Roberts and T. C. Skeat, The Birth of the Codex (London: Oxford University Press for the
British Academy, 1983).
73 The exception is ∏22 (= P. Oxy. 1228), third-century fragments of John 15, now in Glasgow, with
their recto inexplicably blank. It is worth noting that two of the three fragments of the Gospel of
Thomas, P. Oxy. 654 and P. Oxy. 655, are from rolls; P. Oxy. 1 is from a papyrus codex. See B. Layton,
ed., Nag Hammadi Codex II, 2–7 (Leiden: Brill, 1989), pp. 96–7.
 The fourfold Gospel 83
Why did Christians have such a strong predilection for the codex? Skeat
has recently rejected the reasons usually advanced for adoption of the codex
and has insisted that the motive ‘must have been infinitely more powerful
than anything hitherto considered’.74 Skeat notes that the codex could
contain the texts of all four gospels; no roll could do this. He then asks,
‘What can have induced the Church so suddenly, and totally, to abandon
rolls, and substitute not just codices but a single codex containing all four
Gospels?’ He accepts that single gospels circulated as codices, but only as
‘spin-offs’, so to speak, of the four-gospel codex. In his view, the production
of the Fourth Gospel about ad 100 caused a crisis in the church: a formal
decision was taken to publish the four gospels in a single codex, and as a
result the codex became the norm for Christian writings. ‘How the decision
was reached we have no means of knowing. Clearly there must have been
correspondence between the major churches, and perhaps conferences.’75
David Trobisch has recently defended a partly similar theory.76 He claims
that the early Christian use of the codex and of nomina sacra can be ac-
counted for only by positing deliberate decisions concerning the appro-
priate format for canonical writings. Trobisch, however, studiously refrains
from stating when and where this guideline for the preparation of Christian
manuscripts was drawn up.77
The strength of the theories advanced by Skeat and Trobisch is that they
draw attention to the rapid and universal adoption of the codex for what
became in due course canonical writings. However, I am not completely
convinced by either theory. Both theories require a much higher level of
structure and centralized organization within early second-century Chris-
tianity than I think likely.78 If (as Skeat suggests) the four-gospel codex
preceded the circulation of single-gospel codices, it must have been adopted
soon after the beginning of the second century, for we have in ∏52 (usually

74 Skeat, ‘Origin’, pp. 263–8, here 263.

75 Skeat, ibid., p. 263, explains that he no longer accepts either of the two theories which he and C. H.
Roberts had advanced in The Birth of the Codex. For discussion of the various explanations advanced
for Christian adoption of the codex, see J. van Haelst, ‘Les Origines du codex’, in A. Blanchard,
ed., Les Débuts du codex (Turnhout: Brepols, 1989), pp. 13–36; and S. R. Llewelyn, New Documents
Illustrating Early Christianity, Vol. vii (Sydney: The Ancient History Documentary Research Centre,
Macquarie University, 1994), pp. 249–56. T. C. Skeat’s reference to the crisis caused by the publication
of the Fourth Gospel recalls E. J. Goodspeed’s suggestion that the fourfold Gospel emerged a few
years after the appearance of the Fourth Gospel; it was intended to win a wider hearing for the
Gospel of John than it would otherwise have received (Formation, pp. 35–6).
76 Trobisch, Endredaktion. 77 Ibid., especially pp. 12 and 124.
78 I am not convinced by D. Trobisch’s attempt (ibid., pp. 67–8) to overturn this generally held view
of early second-century Christianity.
84 Jesus and Gospel
dated to c. ad 125) a single-gospel codex. But a date soon after the turn
of the century is difficult to square with the ways written gospels and oral
gospel traditions were used at that time.
My own view is that Christian scribes first experimented with single-
gospel codices by adopting the Roman invention of the use of pocket
editions of literary works referred to in ad 84–6 by Martial. In Epigram i.2
Martial recommends that travellers should carry his poems in copies with
small parchment pages which could be held in one hand, presumably parch-
ment codices. Martial even gives his readers the name and address of the
‘publisher’.
Early codices, whether Roman or Christian, were quite small in size
and therefore much more portable than rolls. Christian scribes preparing
writings to be carried by missionaries, messengers, and travellers over long
distances would have readily appreciated the advantages of the codex.79
Their general counter-cultural stance would have made them more willing
than their non-Christian counterparts to break with the almost unanimous
preference for the roll and to experiment with the unfashionable codex.
No doubt the popularity of the codex in Christian circles was enhanced
by its distinctive format. The earliest Christian experiments with the codex
took place at a time when Christians were adopting a distinctive identity
as a tertium genus over against both Judaism and the pagan world. Copying
and using the OT Scriptures and their foundation writings in a new format
was but one of the ways Christians expressed their sense of ‘newness’.80
Some Christians today still cling to a particular translation and format of
the Bible as an identifying mark of their group or theological convictions.
The codex format caught on rapidly in Christian circles.81 The ability
of the codex to hold four gospels seems to have been appreciated at the

79 See especially M. McCormick, ‘The Birth of the Codex and the Apostolic Life-Style’, Scriptorium
39 (1985) 150–8.
80 Colette Sirat rejects the claim made by Roberts and Skeat, Birth, that there was a Jewish origin for
Christian use of the codex. She notes that, in the first centuries of our era, traditional Jewish texts
do not make any allusion to the codex: ‘Le Livre hébreu dans les premiers siècles de nôtre ère: le
témoinage des textes’, in A. Blanchard, ed., Les Débuts du codex, Bildiologia 9 (Brepols: Turnhout,
1989), pp. 115–24.
81 Cf. H. Y. Gamble, Books and Readers in the Early Church: A History of Early Christian Texts (New
Haven and London: Yale University Press, 1995), p. 65: ‘To claim the most primitive edition of
the Pauline letter collection was put out in a codex and that it was the religious authority of Paul’s
collected letters that set the standard for the transcription of subsequent Christian literature in codices
is not to claim that this marked the first use of the codex in Christian circles. It is possible, perhaps
likely, that the codex was first employed in primitive Christianity for collections of texts (testimonia)
from Jewish scripture.’ I accept the latter point, but I am not convinced that it was a collection of
the Pauline letters (which Gamble dates in the early second century, ‘and probably earlier’ (p. 61))
which ‘set the standard’ for use of the codex. We have far more early codices of individual gospels
 The fourfold Gospel 85
very time when four gospels were being brought together in some second-
century circles.
When did this happen? All the evidence I have set out in this chapter
points to the period shortly before 150. Justin’s writings confirm that, in the
decade or so after the Bar Cochba rebellion, Christian self-understanding as
a tertium genus took hold strongly, so perhaps it was during these years that
the four-gospel codex and the fourfold Gospel began to become popular. I
make this suggestion with some hesitation. Numerous pieces of the jig-saw
puzzle are missing; the discovery of only one or two new pieces might well
alter the whole picture.
Acceptance of the fourfold Gospel did not mean the end of oral tradition;
continuing use of oral traditions did not necessarily mean that written
gospels were unknown or of marginal importance. It is a great mistake
to suppose that written traditions and oral traditions were mutually exclu-
sive.82 And it is equally important to note that the emergence of the fourfold
Gospel did not instantly suppress either the use of, or the production of,
further gospels. To have a set of four authoritative gospels does not mean
that one stops reading anything else. In some circles doubts emerged from
time to time about one or more of the four gospels; the universal adoption
of a four-gospel canon took much longer. Above all, we need to recall that,
even in Irenaeus’ day, when the fourfold Gospel was axiomatic in many
circles, the sayings of Jesus possessed an even higher authority.
It will be clear from the preceding paragraphs that I prefer a theory of
gradual development to a ‘big bang’ theory. I envisage the following stages
in the emergence of the fourfold Gospel, though of course I recognize that
my summary suggests a neat and tidy development, far removed from the
reality of continuing debate and of diversity of practice. The codex began
to be used for individual gospels soon after the turn of the century, a period
when a plurality of gospels was known in many circles. During these decades
the term ‘Gospel’ was used both for oral and for written ‘Jesus’ traditions.
Use of the term ‘Gospel’ for two or more writings raised the question of their
relationship to the one Gospel about Jesus Christ. This problem was solved

and codices of the four gospels than we do of collections of Paul’s letters; the gospels are quoted
much more frequently in the second century than are the Pauline epistles. As I have noted above,
early Christian use of the LXX in codices must also have encouraged Christians to adopt the codex
as a standard format. For further discussion, see Chapter 8 below.
82 See Gamble, Books and Readers, pp. 28–30; G. N. Stanton, ‘Form Criticism Revisited’, in M. D.
Hooker and C. J. A. Hickling, eds., What about the New Testament? FS C. F. Evans (London: SCM,
1975), pp. 13–27. See also Loveday Alexander, ‘The Living Voice: Scepticism towards the Written
Word in Early Christian and in Graeco-Roman Texts’, in D. J. A. Clines, S. E. Fowl, and S. E.
Porter, eds., The Bible in Three Dimensions (Sheffield: JSOT, 1990), pp. 221–47.
86 Jesus and Gospel
by use of the title eÉagg”lion kat† . . . for individual gospels. The use of
this title facilitated both the acceptance of the fourfold Gospel and the use
of the codex for four gospels. The four-gospel codex strongly encouraged
acceptance of the fourfold Gospel, and vice versa: both are likely to have
taken place for the first time shortly before the middle of the second century.
The fourfold Gospel did not gain immediate acceptance: as I shall recall
in a moment, there were very strong currents running in the opposite di-
rection. Christian insistence that the church had four equally authoritative
stories written by apostles and their followers left doors wide open for both
Jewish and pagan critics. The continuing attraction for many Christians
and ‘heretics’ of one written Gospel (whether or not a harmony), as well
as the jibes of critics, encouraged Irenaeus to mount what seems to have
been the first full theological defence of the fourfold Gospel. The universal
acceptance of four gospels – a four gospel canon – followed in due course,
but not without further vicissitudes.

3.6 theological significance

The acceptance of the fourfold Gospel does not raise problems for the
historian. On the contrary, the historian is happy to have four sources
available for reconstruction of the actions, words, and intentions of Jesus. Of
course, the historian will want to assess all sources critically, but, in principle,
the more sources the better. However, as soon as the historian considers the
broad sweep of the development of the early Christian movement in the
period before Constantine, it becomes clear that the four canonical gospels
played a greater part in the development of the Christian movement and
of Christian theology than any non-canonical Jesus or gospel traditions.
The fourfold Gospel is of no significance if one wants to make a parti-
cular historical reconstruction of the life of Jesus the sole focus of religious
concern. For example, if one really believes that nothing more can be said
about Jesus than that he was a Wisdom or a Cynic teacher, then one is opting
for a post-Christian position: in that case, the four evangelists’ portraits of
Jesus will be of no more than passing historical interest, to be discarded
along with numerous later faith images of Jesus. But, if one accepts that the
evangelists’ attempts to tell the story of Jesus and to spell out his significance
are in some way normative for Christian faith, then the fourfold Gospel is
problematic. Why do Christians need four stories?
The momentous nature of the decision to accept four gospels becomes
clear once we recognize that alternative solutions nearly won the day in
the first and second centuries. Acceptance and use of one gospel was a live
 The fourfold Gospel 87
option again and again. When Matthew wrote his gospel, he did not intend
to supplement Mark: his incorporation of most of Mark’s Gospel is surely
an indication that he intended that his gospel should replace Mark’s, and
that it should become the Gospel for Christians of his day. Similarly Luke.
Luke’s Preface should not be dismissed merely as the evangelist’s way of
honouring literary convention. There is little doubt that Luke expects that
his more complete gospel will displace his predecessors’, even though he
may not intend to disparage their earlier efforts. Whether or not John knew
of the existence of one or more of the synoptic gospels, he seems to have
expected that his gospel would win wide acceptance as the Gospel.
This pattern continued in the second century. Numerous very different
Christian groups used only one gospel. Immediately before his declaration
that there can be neither more nor less than four gospels, Irenaeus notes
that four groups of heretics, Marcion included, have all fastened on one
of the four gospels (Adv. Haer. iii.11.7). Irenaeus is painting with plenty
of rhetoric on his brush at this point; nonetheless, his picture is broadly
accurate.
And then there is Tatian, who produced one harmonized gospel out of
four, or possibly five, gospels. He was almost certainly not the only person
to opt for this solution, and his solution was amazingly successful in some
circles for a very long time.
In the first and second centuries there were strong tides moving Christian
churches towards acceptance of only one gospel. Acceptance of the fourfold
Gospel meant turning back the tidal currents: it carried as a corollary
rejection of all the various attempts to opt either for one single gospel,
or for a harmony of known gospels. The tide was not turned back in a
hurry, but, once it began to turn, there was no going back: we never find
manuscript evidence for acceptance of any ‘fifth’ gospel such as Thomas
or Peter alongside one or more of the writings in the fourfold Gospel.83
The adoption of the four-gospel codex undoubtedly encouraged this whole
process.

Was Irenaeus correct to give such a robust extended theological defence of

the fourfold Gospel? Was he wise to give it a privileged theological position?
Most of us have difficulty in accepting his view that two of the evangelists
were apostles, and two were close associates of apostles. However, a careful
reading of his writings reveals that his notion of ‘apostolic’ is acceptably
broad, for Irenaeus emphasizes the lines of continuity of the four written

83 So too, J. K. Elliott, ‘Manuscripts, the Codex and the Canon’, JSNT 63 (1996) 107.
88 Jesus and Gospel
gospels with the oral Gospel proclaimed by the apostles; like the apostles,
the four gospels all proclaim one God, one Christ; they are held together
by one Spirit.
At about the same time, Serapion, bishop of Antioch, made a similar
theological judgement. The Gospel of Peter should not be accepted solely
because the great apostle’s name was attached to it: continuity with the
apostolic faith was the criterion by which it should be judged. So too Luther
in his insistence that the test of apostolicity was whether or not a book
proclaimed Christ. ‘That which does not preach Christ is not apostolic,
though it be the work of Peter or Paul, and conversely that which does
teach Christ is apostolic even though it be written by Judas, Annas, Pilate,
Herod.’84
Once we understand ‘apostolic’ in this extended sense, we need not
hesitate to affirm Irenaeus’ defence of the fourfold Gospel. The other gospels
which Irenaeus knew, or which appeared after his day, are clearly beyond the
limits of acceptable theological diversity at crucial points. For Irenaeus there
were three crucial theological points: the doctrine of one God, the Creator
of all; continuity with the Scriptures; and Christology. If we consider all
the possible rivals to the four gospels which became canonical, they all fall
down on one or more of these theological criteria.
At one point Irenaeus attacks the Valentinians for audaciously accept-
ing ‘The Gospel of Truth’. He notes that it is totally unlike the gospels of
the apostles, and also that it is a comparatively recent writing (Adv. Haer.
iii.11.9). In modern times some weight has often been attached to this crite-
rion of ‘earliness’: the four gospels are authoritative for Christian theology
because they are the earliest witnesses we have to the actions and teaching
of Jesus. Caution is of course necessary, for there are ‘non-canonical’ tra-
ditions which have a good claim to be as early as traditions which found
their way into the four ‘canonical’ gospels. But their importance must not
be exaggerated. Even the Jesus Seminar accepts as authentic only five of
the logia of Thomas which are not found in the canonical four.85 And,
as for complete gospels, are any earlier than the canonical four? Surely
J. D. Crossan is exercising a vivid historical imagination when he claims
that an early version of the Gospel of Peter was written in the fifties, possibly
in Sepphoris.86

84 M. Luther, Werke (Erlangen, 1826–57), pp. 63, 156f. I owe this reference to R. H. Bainton in the
Cambridge History of the Bible, Vol. iii, ed. S. L. Greenslade (Cambridge: Cambridge University
Press, 1963), p. 7.
85 See further Stanton, Gospel Truth? (above, n. 37), pp. 84–93. 86 See further ibid., pp. 78–82.
 The fourfold Gospel 89
In short, a critical Christian theology need not stumble over the fourfold
Gospel. But there are corollaries, four of which I shall single out for brief
comment. First, the question of genre. If the four gospels are regarded
primarily as theological witnesses to Jesus Christ in narrative form, then
it makes good sense to retain these four primary witnesses. But, if the
four are considered primarily as historical records, then along with Tatian
it would surely be preferable to roll up the four into one. I think that
it is clear from the brilliant detective work of Tjitze Baarda and William
Petersen that Tatian’s concerns were primarily historical and included a
quest for unity and harmony.87 By accepting the fourfold Gospel, the early
church acknowledged that the gospels are not histories; if we follow suit,
we are accepting them as witnesses in narrative form, in spite of their
discrepancies and contradictions. They belong to the broad genre of b©oi
(lives), but they are not b©oi tout court; they are four witnesses to the one
Gospel.
Secondly, the fourfold Gospel has major implications for Christology.
I have drawn attention to the way both the Muratorian Fragment and
Irenaeus insisted that, in spite of the different beginnings of the gospels, they
were held together by one Spirit. Their comments may have been fuelled
by pagan attacks on inconsistencies in the openings of the gospels, but
their fascination with the beginnings of the gospels is surely an indication
of awareness of the different Christological perspectives which result. This
was seen clearly by Theodore of Mopsuestia at the end of the fourth century.
He also commented on the different beginnings of the gospels, and noted
that in the synoptic gospels teaching on the divinity of Christ was almost
entirely lacking: that is why John opened his Gospel with an immediate
reference to the divinity of Christ.88
Acceptance of the fourfold Gospel carries with it a commitment to the
Christological tension between the synoptics and John. The history of
Christological discussion right up to the present day reminds us again and
again that we ignore this creative tension in perspective, at our peril. The
terminology changes – Christology from below and above, implicit and
explicit Christology – but the fundamental Christological issue is marked
out by the fourfold Gospel: the very different Christological stances of the

87 Baarda, ‘DIAFWNIA – SUMFWNIA’ (above, n. 27); W. L. Petersen, ‘Textual Evidence’ (above,

n. 51), and his Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in Scholarship
(Leiden: Brill, 1994).
88 Theodore of Mopsuestia, Commentarius in Evangelium Iohannis Apostoli, Corpus Scriptorum Chris-
tianorum Orientalium 116 (Paris and Louvain, 1940). I owe this reference to H. Merkel, Die Pluralität
(above, n. 26), §36.
90 Jesus and Gospel
synoptics and the Fourth Gospel should not be blurred, but should both
be taken with the utmost seriousness.89
Thirdly, the fourfold Gospel forces us to reflect on a range of hermeneu-
tical issues. By accepting the fourfold Gospel, we are ignoring the intention
of two, or possibly three, of the evangelists, and in a sense we are encour-
aging the four gospels to interpret one another. These observations do not
lead me, at least, to abandon the search for the original intention of the
evangelists. But they do remind me that it is the fourfold Gospel which
has fed the life of the church for nearly two thousand years, not any one or
more single gospels. I personally do not want to give hermeneutical priority
to the canonical shape of the gospels over against the original intention of
the evangelists: I want to try to take both seriously and critically, along with
the ways the gospels have been understood within the continuing Christian
tradition.
By accepting that the four are all witnesses to the one Gospel, we are
forced to reflect both on the theological convictions which they share and
on the points at which they diverge. Why do we prefer one evangelist’s
portrait or emphasis? What are our theological criteria for making such
judgements? Has the continuing scholarly interest in the distinctive features
of the individual gospels blinded us to the theological concerns they have
in common?
Fourthly, the fourfold Gospel raises liturgical issues, especially for con-
temporary compilers of lectionaries. Most modern lectionaries treat the
gospels as individual writings: one year of lectionary readings is devoted
to each gospel. Most churches in the United Kingdom have adopted a
three-year cycle in which a whole year is devoted to each of the synoptic
gospels, with passages from John’s Gospel inserted at appropriate points. Is
this compromise satisfactory? I think that it probably is, though extended
use of the new pattern may prove otherwise.

I end with my beginning, with Irenaeus. I cannot accept some of the reasons
he offers in defence of the fourfold Gospel. But I do accept his theological
conviction that the fourfold Gospel is the pillar which sustains the church.
This is a static image, so perhaps some may prefer the image which ap-
pealed to Hippolytus, Cyprian, Victorinus of Pettau, and, very recently, to
Rudolf Schnackenburg: the four gospels are like the rivers of Paradise which
flow from the Garden of Eden into the whole known earth at that time

89 On this point, see especially Morgan, ‘Hermeneutical Significance’ (above, n. 6), pp. 380–6.
 The fourfold Gospel 91
(Gen. 2.10-14).90 The Biblical image of rivers and living, flowing waters
was often linked to the gift of the Spirit, as it was by the Fourth Evangelist.
So perhaps Irenaeus, with his emphasis on the Spirit who holds together
the fourfold Gospel, would have been happy with this later, more dynamic
image. After all, for Irenaeus the foundation and pillar of the church is the
fourfold Gospel and the Spirit of life (stÓlov d• kaª stžrigma –kklhs©av
t¼ eÉagg”lion kaª PneÓma xw¦v, iii.11.8).91

90 R. Schnackenburg, Jesus in the Gospels: A Biblical Christology (Louisville: Westminster John Knox,
E. tr. 1995), pp. 324–5. Schnackenburg refers to Gen. 2.10-14, but does not note that this image also
appealed to the patristic writers listed above. See Merkel, Die Widersprüche, p. 7 n. 1, who gives full
references.
91 I have quoted the Greek text from fragment 11 (Anastasius Sinaita), as edited by Rousseau and
Doutreleau, Irénée (above, n. 11), pp. 160–2.
 chap t e r 4

Jesus traditions and gospels in Justin Martyr

and Irenaeus

The status of Jesus traditions and of the ‘canonical’ gospels gradually grew
in the course of the second century.1 At the beginning of the century there
was widespread respect for ‘words of the Lord’ and for ‘the Gospel’ (whether
oral or written) in which Jesus traditions were embedded. By the end of
the century the early church seemed to be within a whisker of accepting a
‘canon’ of four written gospels, no more, no less.
I do not intend to discuss all the developments and factors which led
to the sea change which took place during the second century. In order to
do so I would need to offer many hostages to fortune, for at crucial points
the evidence is disputed, particularly with reference to the first half of the
second century. For example, although the Didache has usually been dated
to the first decades of the second century, it is now generally accepted that
it contains several layers of traditions, the dating of which is problematic.
A major challenge has been mounted to the consensus that Ignatius wrote
seven letters in the early years of the second century. I do not think that the
challenge is likely to be successful, but discussion of it would be a distraction
from my primary task. And do we know the date of II Clement?
I shall focus my attention on two second-century giants whose substantial
surviving writings can be dated with some confidence, Justin Martyr and
Irenaeus. There is general agreement that their writings are important for
my topic – and for many others. It will be my contention that some of
their evidence for the status and use of Jesus traditions and of gospels has
been misconstrued or overlooked in recent discussion. All too often their
writings have been subject to what I call ‘cherry picking’: tasty morsels have
been plucked in order to garnish a grand theory, often at the expense of
a close reading of the texts. As we shall see, there has been no shortage of
grand theories.
1 An earlier version of this chapter was given as a main lecture at the Jubilee Meeting of the Colloquium
Biblicum Lovaniense in July 2001. It is published in J.-M. Auwers and H. J. de Jonge, eds., The Biblical
Canons (Leuven: Peeters, 2003), pp. 351–68.

92
 Justin Martyr and Irenaeus 93
I shall not attempt to discuss in any detail the text form or the source of
the Jesus traditions. It is important to try to establish whether quoted Jesus
traditions are from this gospel or that, from oral or from written sources,
from a pre- or a post-synoptic harmony. These questions have received
plenty of scholarly attention. I shall not avoid these fascinating issues, but I
shall concentrate on a different, somewhat neglected, set of questions. What
status do Justin and Irenaeus give to Jesus traditions and to the gospels?
Are they merely respected traditions? Are they cited as authoritative texts?
Are they considered to be Scripture? Do they have the same standing in the
church as the OT writings? How close are we to the later emergence of the
concept of a canon, an agreed list of authoritative writings which cannot
be altered?

4.1 justin mart yr

Justin’s comments on the Septuagint and the alleged Jewish deletions and
modifications to the text of what Justin regards as Scripture have often been
noted.2 The hermeneutical principles which underlie Justin’s thorough
Christianization of the Old Testament have been discussed less frequently,
but they are of considerable importance for Christian theology. Both topics
are fascinating. However, the key question for this chapter lies elsewhere:
where are we to place Justin in the story of the eventual emergence of the
NT canon?
Is Justin, as C. H. Cosgrove has claimed, a reactionary figure who stands
four square against second-century trends towards regarding apostolic writ-
ings as canon?3 Can we accept Cosgrove’s claim that Justin even devalues
the authority of the emerging NT canon, limiting himself to the sayings of
Jesus, with misgivings about the emerging canonical status of the gospels?4

2 We now have available Miroslav Marcovich’s much-needed critical editions of the Greek text, Iustini
Martyris Apologiae Pro Christianis, Patristische Texte und Studien 38 (Berlin: De Gruyter, 1994) and
Iustini Martyris Dialogus cum Tryphone, Patristische Texte und Studien 47 (Berlin: De Gruyter, 1997).
Marcovich regularly proposes additions and corrections to the Parisinus codex, the one surviving
manuscript of any importance; this huge codex is dated 11 September 1363. Although Marcovich’s
own editorial proposals may be a touch too radical for some, they are indicated very clearly in his
printed text and can be readily ignored if necessary; details of the editorial emendments made by his
predecessors are also indicated. Marcovich’s editions provide a solid platform for fresh study of these
fascinating writings; nonetheless, they remind us that the text of Justin’s writings is in a parlous state.
3 C. H. Cosgrove, ‘Justin Martyr and the Emerging Christian Canon: Observations on the Purpose and
Destination of the Dialogue with Trypho’, Vig. Chr. 36 (1982) 209–32. There are several weaknesses
to Cosgrove’s case, the most important of which is his failure to consider the evidence of Justin’s
Apologies.
4 See Cosgrove, ‘Justin Martyr’, 226–7.
94 Jesus and Gospel
Or does Justin have a very high regard for the sayings of Jesus and the
gospels and consider them to be as authoritative as the OT writings?
At first sight Justin’s writings seem to offer limited evidence for our quest,
for only once does he refer to an individual NT writing by name.5 However,
as in the interpretation of any writing, whether ancient or modern, genre
and context must not be ignored. Justin’s Apologies and his Dialogue with
Trypho are apologetic writings, which address two very different readerships.
Perhaps Justin does hope that his Apologies will win the Emperor Marcus
Aurelius and the leading Gentile opinion formers of his day over to the
Christian ‘philosophy’. Perhaps he does hope that Trypho and other Jewish
teachers will acknowledge that Jesus is the Messiah promised in Scripture.
In my view, however, it is more likely that Justin’s extant writings were
all intended to provide Christian members of his philosophical school in
Rome with apologetic material which they could use in their encounters
with both Jews and Gentiles.
In either case, reference to the names of the authors of the NT writings
would not have been appropriate. As J. B. Lightfoot noted, ‘In works like
these, addressed to Heathens and Jews, who attributed no authority to
the writings of Apostles and Evangelists, and for whom the names of the
writers would have no meaning, we are not surprised that he refers to those
writings for the most part anonymously and with reserve.’6
Justin’s writings provide us with plenty of evidence to assess, though that
task is not easy.7 The First Apology was written very shortly after ad 150, the
Dialogue with Trypho only a few years later in the same decade. However,
it is unwise to try to trace development in Justin’s thinking. Some sections
of the Dialogue may well have been written before the Apology and in-
serted into that long, rather rambling account of Justin’s conversations with
Trypho.8

5 The Revelation of John (Dialogue 81.4).

6 Essays on the Work Entitled Supernatural Religion (London, 1893), p. 33. I owe this reference to Charles
E. Hill, ‘Justin and the New Testament Writings’, in E. A. Livingstone, ed., Studia Patristica 30
(Leuven: Peeters, 1997), p. 43.
7 E. F. Osborn, Justin Martyr (Tübingen: Mohr, 1973), p. 120 notes that in 1877 B. L. Gildersleeve
claimed that ‘the battle over the question whether Justin’s Memoirs of the Apostles are identi-
cal with our canonical gospels has lasted nearly a century. Begun by Stroth in 1777, it is safe to
say that the fight is going on at this very moment in the powder magazine of some theological
review.’
8 For discussion of Justin’s possible use of earlier sources or his own earlier compositions, see O.
Skarsaune, The Proof from Prophecy. A Study in Justin Martyr’s Proof-Text Tradition: Text-Type, Prove-
nance, Theological Profile (Leiden: Brill, 1987), and Luise Abramowski, ‘Die “Erinnerungen der
Apostel” bei Justin’, in P. Stuhlmacher, ed., Das Evangelium und die Evangelien (Tübingen: Mohr,
1983), pp. 341–54.
 Justin Martyr and Irenaeus 95
4.1.1 Jesus traditions: ‘the words of the Saviour’
The opening chapters of the First Apology underline the importance Justin
attaches to the teachings of Christ and to their careful transmission (4.7;
6.2; 8.3). In all three passages Justin emphasizes that Christians teach (or
hand over, parad©dwmi) what they have been taught by Christ, whom they
‘worship and adore’ along with ‘the most true God’ (6.1). The teachings
of Christ are clearly authoritative, but nothing further is said about their
status in these chapters.
At the climax of the Apology, however, three related passages leave little
doubt that the carefully transmitted traditions Justin refers to include the
written memoirs of the apostles. The prayer following the baptism of ‘the
one who has been illuminated’ asks that Christians show by their deeds
that they are ‘good citizens and guardians of what has been commanded’
(65.5, –nt”llomai). Only those who ‘live as Christ handed down’ (66.1,
parad©dwmi) are permitted to participate in the baptismal eucharist. The
food ‘eucharistized’ through ‘the word of prayer’ that is from Jesus Christ
is life-giving (66.2). There follows a much-discussed passage to which we
shall return: ‘For in the memoirs composed by them, called gospels, the
apostles handed down what they had been commanded: that Jesus took bread
and having given thanks said . . . This is my body’ (66.3, par”dwkan
–ntet†lqai aÉto±v). Justin is adamant that traditions of the sayings and
actions of Jesus have been transmitted carefully in the written memoirs by
the apostles, and are handed on and carried out by his fellow-Christians.
Their ultimate source is Jesus Christ himself. Their authoritative status
could hardly be underlined more firmly, even though they are not referred
to as ‘Scripture’.
In chapter 14 of the Apology Justin provides an important full introduc-
tion to no fewer than twenty-six sayings of Jesus, organized topically into
ten groups.9 Christians ‘pray for their enemies’ and try to persuade those
who hate them unjustly ‘to live according to the good suggestions of Christ’
(14.4). Justin then states that he is about to cite some of the teachings given
by Christ. He makes two comments on their character.
(a) They are ‘short and concise sayings’ (brace±v d• kaª sÅntomoi parì
aÉtoÓ l»goi), for Christ was no sophist. Justin assumes that his addressees
will be familiar with long-winded sophists and thus impressed by the

9 A. J. Bellinzoni, The Sayings of Jesus in the Writings of Justin Martyr, SNT 17 (Leiden: Brill, 1967),
pp. 49–100, examines in detail the text form and source of these sayings. However, he fails to comment
on Justin’s important introductory remarks in chapter 14.
96 Jesus and Gospel
pithy sayings of Jesus. In an effort to impress Trypho, the sayings of Jesus
are also said to be ‘short’ (brac”a l»gia) at Dialogue 18.1.
(b) Justin claims that the ‘commandments’ (dogm†ta, 14.4) which are
about to be quoted are the word of Christ, and ‘his word was the power of
God’ (14.5, dÅnamiv qeoÓ ¾ l»gov aÉtoÓ §n). This introductory comment
is clearly intended to establish the importance and authoritative status of
the twenty-six sayings of Jesus which follow in chapters 15–17.
Justin himself opens each set of sayings by announcing its theme; the
sayings are then introduced with simple phrases in the aorist tense: ‘he said’,
‘he taught’, ‘he commanded’. For example, ‘Concerning chastity, he said
this . . .’; this introduction is followed by four sayings linked only by ka©
(15.1-4). The fourth set of sayings is introduced by Justin as follows: ‘And
that we should share with the needy . . . he said these things . . .’; eight
sayings linked only twice with ka© are juxtaposed (15.10-17).
This pattern is repeated almost identically ten times over. The penul-
timate paragraph is the single exception (17.1-2). Here Justin insists that
Christians pay taxes ‘as we have been taught by him’ and then sets out
a much-abbreviated version of the pronouncement story concerning pay-
ment of tribute to Caesar, Mark 12.13-17 and parallels.
The ten sets of Jesus traditions are linked to their present context very
loosely. With the partial exception of Dialogue 35.3, where Justin cites one
set of four sayings concerning false teachers and false prophets, there are no
comparable passages in his writings. These ten sets of sayings of Jesus were
almost certainly collected and arranged by Justin himself for catechetical
purposes in his school in Rome. Their status is underlined by Justin’s
introduction of them as ‘the power of God’ and by Justin’s repeated claim
that his fellow-Christians transmit, and live by, the sayings of Jesus which
have been carefully handed on to them. For our present purposes, we may
leave it open whether the twenty-six sayings of Jesus quoted in chapters 15–
17 have been taken from oral or from written sources.
From the First Apology we turn to the Dialogue. Here Justin includes
explicit statements concerning the sayings of Jesus in his account of his
conversion to Christianity, which he dubs ‘philosophy safe and simple’ (8.1).
He tells his Jewish opponent Trypho that at the time of his conversion he
experienced a passionate desire for the prophets, and for those men who
are the friends of Christ, presumably the apostles. He then expresses the
hope that all people should be as keen as he is not to distance themselves
from the Saviour’s words (mŸ ˆf©stasqai tän toÓ swt¦rov l»gwn), for
they evoke profound awe (d”ov). Their innate power puts to shame those
who turn aside from the right way, while pleasant rest (ˆn†pausiv) comes
 Justin Martyr and Irenaeus 97
to those who carry them out (8.2). Here a version of the ‘two ways’ ethical
tradition is linked to the effect the dynamic sayings of Jesus have.
In his vigorous reply Trypho claims that in his conversion to Christianity
Justin has been led astray by false statements and has followed men who
are not at all worthy. From the context, the latter can only be ‘the friends
of Christ’, i.e. the apostles. Trypho soon concedes that he has taken some
trouble to read the admirable and great commands of Christ ‘in the so-called
Gospel’ (10.2), a point Justin repeats at 18.1. At Dialogue 88.3, shortly before
the thirteen references to ‘the memoirs of the apostles’ in chapters 100–7,
Justin states that ‘the apostles of this our Christ’ have written that the Spirit
fluttered down on Jesus at his baptism like a dove. This is clearly a reference
to at least two apostles’ writings.
So, long before the thirteen references in the Dialogue to ‘the memoirs
of the apostles’ are reached in chapters 100–7, the reader can hardly avoid
the conclusion that the powerful words of the Saviour may be read in the
writings of the apostles, the friends of Christ.
This key point emerges again in a particularly dramatic passage in
Dialogue 113 and 114. Justin states that the words of Jesus are the sharp
knives by which Gentile Christians in his own day have experienced the
‘second circumcision’ (of their hearts). The first physical circumcision is for
Jews. The second circumcision, ‘which circumcises us from idolatry, and
in fact all vice’, is carried out ‘by the words spoken by the apostles of the
Corner Stone’ (Dialogue 113.6-7; 114.4). Given the preceding references to
reading and writing, we can be all but certain that written traditions are
in mind here. The authoritative status of those traditions could hardly be
underlined more strongly.
How does Justin understand the relationship of the sayings of Jesus
transmitted through the writings of the apostles to Scripture? In I Apol.
61-3, sayings of Jesus are set alongside cited words of Isaiah, with the clear
implication that they have the same status. The words of Christ, ‘Unless
you are born again . . .’, are followed almost immediately by a version of
Isa. 1.16-20, ‘Thus spoke Isaiah the prophet’ (61.4-8). The introductory
formulae are almost identical: ‘Christ said’ and ‘Isaiah the prophet thus
spoke.’
In the next chapter the burning-bush theophany is Christianized: ‘our
Christ’ converses with Moses in the form of fire out of the bush and said,
‘Unloose your sandals and come near and hear’ (62.3). This is followed by
a citation of Isa. 1.3, introduced as words of the prophetic Spirit through
Isaiah the prophet (63.1). Justin then cites two sayings of Jesus, introduced
as ‘Jesus Christ said’ and ‘Our Lord himself said’ (63.3-5).
98 Jesus and Gospel
The first readers of the Apology are encouraged to conclude that sayings
of Jesus have the same status as the words of Isaiah, though they are left
to draw this conclusion for themselves. Justin makes this key point more
explicitly in the Dialogue. With a rather quaint rhetorical touch, Justin says
to his Jewish opponent Trypho, ‘Since you have read what our Saviour
taught, as you have yourself acknowledged, I think I have not acted in an
unseemly fashion by adding some short sayings of his [Christ’s] to those
found in the prophets’ (18.1). The immediate context is significant. In the
preceding chapter three passages from Isaiah are linked (52.5; 3.9-11; 5.18-
20) as a preface to three sayings of Jesus. The first of the latter sayings is
a version of Matt. 21.13, words of Jesus addressed to the money-changers
in the temple. Jesus refers to Jer. 7.11 with the words, ‘it is written: “My
house is a house of prayer, but you have made it a den of robbers.”’ Justin’s
version is much closer to Matt. 21.13 than to the LXX (Dialogue 17.3), so in
all probability Justin has used Matthew’s Gospel at this point.
There is a further example of the use of g”graptai in a quotation from
Matthew at Dialogue 78.1. Here the quotation of Mic. 5.2 at Matt. 2.5 is
referred to. One might have expected Justin to have imitated this NT usage
in his own introductions to some of the sayings of Jesus he cites, thus
placing them on all fours with the OT passages he quotes so frequently.
However, although Justin is familiar with the term grafž for Scripture
(e.g. Dialogue 56.12,17), he never uses g”graptai to introduce either an
OT quotation or a saying of Jesus as Scripture.10
Nonetheless, it would be a grave mistake to claim that the sayings of
Jesus are in any way inferior to Scripture. This emerges very clearly from
Dialogue 119.6. ‘For as he [Abraham] believed the voice of God, and it was
imputed to him for righteousness, in like manner we, having believed God’s
voice spoken through the apostles of Christ, and preached to us through
the prophets, have renounced even to death all the things of the world.’
Charles Hill comments appropriately: ‘Here it is God’s own “voice” which
has spoken “through” the apostles of Christ, and here Justin explicitly and
boldly places this “inspiration” on a par with that of the OT prophets.’11
Justin attaches considerable significance to the teachings of Christ. Their
authority and power are clear. Justin’s readers are left in no doubt that the

10 H. von Campenhausen, The Formation of the Christian Bible (Philadelphia: Fortress, E. tr. 1972),
p. 170, notes that in the ‘later anti-Jewish Dialogue’ Justin introduces a dominical saying with the
words ‘it is written’, and suggests cautiously that this is Justin’s way of referring to texts which are to
be acknowledged as authentic and normative. I do not think that the three uses of g”graptai in
question (49.5; 100.1; 105.6) bear this weight.
11 Hill, ‘Justin and the New Testament Writings’, p. 48.
 Justin Martyr and Irenaeus 99
sayings of Jesus have the same standing as the words of the prophets. Justin
returns to this point in the closing pages of the Dialogue. With an ironical
touch he rounds on Trypho: ‘If the teaching (did†gmata) of the prophets
and of Christ disquiet you, it is better for you to follow God rather than
your unintelligent and blind teachers’ (134.1). As we have noted above, in
several passages Justin remarks that ‘the words of the Saviour’ have been
conveyed in the writings of the apostles. So it is appropriate that we should
now consider the status Justin attaches to ‘the memoirs of the apostles’.

4.1.2 The gospels: ‘the memoirs of the apostles’

At the climax of the First Apology Justin refers twice to ‘the memoirs of the
apostles’ (t‡ ˆpomnhmoneÅmata tän ˆpost»lwn, 66.3; 67.3) in contexts
which underline their importance; the phrase is used a further thirteen
times in one section of the Dialogue (chs. 100–17).
In his account of the origin and significance of the Christian eucharist,
Justin quotes Jesus’ words of institution as they are recorded in the memoirs
written by the apostles ‘which are called gospels’ ( kale±tai EÉagg”lia,
First Apology 66.3). The status of the memoirs could hardly be clearer. But is
the explanatory reference to the gospels a later addition to the text? This is
the only time Justin refers to the noun eÉagg”lion in the Apologies, so it
is not surprising that some scholars have claimed that the clause is a later
gloss. We cannot ignore the fact that this is the first time the plural ‘gospels’
is used in early Christian writings. Even a generation after Justin, Irenaeus
only rarely used the phrase ‘gospels’ in the plural; he much preferred ‘the
gospel’, or ‘the Gospel according to . . .’12
On the other hand, Justin is in the habit of adding explanatory phrases or
clauses, especially (as here) when his putative readers may have been baffled
by his terminology.13 So, with this one exception, Justin may have felt that
it was quite unnecessary to explain that ‘the memoirs of the apostles’ were
‘gospels’. A decision is difficult, but I have already given plenty of evidence
(and more will follow) to establish that the ‘memoirs’ were the writings
which became known as ‘the gospels’.
The second and final reference in the First Apology to the memoirs of the
apostles occurs in Justin’s account of what takes place in his own day in the
Sunday gathering of Christians. ‘The memoirs of the apostles or the writings
12 See A. Benoit, Saint Irenée. Introduction à l’étude de sa théologie (Paris: Presses Universitaires de
France, 1960), pp. 103–50.
13 See further Abramowski, ‘Die “Erinnerungen” ’, p. 323; I Apology 65.1; 65.5 and Dialogue 10.2 are
noted as parallels to this explanatory clause.
100 Jesus and Gospel
of the prophets are read, for as long as time permits’ (67.3, m”criv –gcwre±).
This is the earliest extant reference to the reading of ‘the gospels’ in the
context of Christian worship. The reading is followed by an exposition given
by ‘the Ruler’ (¾ proestÛv). There is no lectionary; the only constraint
on the length of the passage read is the time available. The ‘memoirs of
the apostles’ are considered to be of equal importance to ‘the writings of
the prophets’. Indeed, they may even be given a measure of precedence
by being referred to before the prophets.14 The liturgical setting provides
further evidence that status of the memoirs was very high indeed.
How many gospels does Justin know? Justin uses the noun eÉagg”lion
twice in the singular in the Dialogue. At 10.2 Trypho is allowed to express
admiration for the precepts of Christ recorded ‘in the so-called gospel’ (–n
t
legom”nw„ EÉaggel©w„); he has even taken the trouble to read them!
Justin introduces his version of Matt. 11.27 = Luke 10.22 with ‘it is written
in the gospel, saying . . .’ (100.2). In both cases, as in several other early
Christian writings, the singular is to refer to the ‘one gospel’ in which
sayings of Jesus are written.
The phrase ‘the memoirs of the apostles’ could be taken to refer to the ‘one
gospel’, but this is most unlikely. We have already noted the explanatory
clause in the First Apology 66.3, ‘the memoirs which are called gospels’, and
the important comment at Dialogue 88.3: ‘the apostles of this our Christ’
have written that the Spirit fluttered down on Jesus at his baptism like a
dove.
Confirmation that in his references to ‘the memoirs of the apostles’ Justin
has in mind more than one written gospel is provided by two of the thirteen
references to the memoirs in the Dialogue. At 103.8 an explanatory clause
follows a reference to the memoirs: ‘which, I say, were composed by his
apostles and those who followed them’. This comment on the composition
of the memoirs implies that they were written by more than one apostle,
and more than one follower of an apostle, i.e. Justin accepts at least four
gospels, though, unlike Irenaeus, he does not name them or discuss their
differences.15 It is a natural, but not a necessary, inference that Justin has in
mind gospels written by the apostles Matthew and John, and by followers

14 Although it has sometimes been suggested that the prophets are Christian prophets, this is unlikely,
given Justin’s repeated references to, and respect for, the OT prophets. The possibility of choice
between a reading from the gospels or (¢) a reading from the prophets is puzzling. Since this is out
of line with later liturgical practice, the text is unlikely to be faulty.
15 Osborn, Justin Martyr, seems to have missed this important passage. He notes that t‡
ˆpomnhmoneÅmata may have a singular meaning. ‘If plurality of authorship were important, some
further description of the apostles and their writings could be expected’ (p. 124).
 Justin Martyr and Irenaeus 101
of the apostles, Mark and Luke. However, caution is necessary. As we shall
see in a moment, it is possible that Justin considered Mark’s Gospel to be
one of the memoirs of the apostles, i.e. to stem from Peter; and we cannot
be confident that Justin had John’s Gospel in mind when he penned the
phrase ‘memoirs composed by his apostles and those who followed them’.
Dialogue 106.3 is more problematic: ‘We are told that he [Christ] changed
the name of one of the apostles to Peter, and it is written in his memoirs (–n
to±v ˆpomnhmoneÅmasin aÉtoÓ) that this took place . . .’. Whose memoirs
are referred to here? If we take without emendation the text of the sole
witness, the fourteenth-century Parisinus codex, there are two possibilities.
The memoirs could be Christ’s or Peter’s. Justin does not refer elsewhere
to the memoirs of one individual; only once does he ever name the author
of any earlier Christian writing (the Revelation of John, Dialogue 81.4).
Nonetheless, both Zahn and Harnack interpreted this sentence as a refer-
ence to Peter’s memoirs, i.e. to Mark’s Gospel, as does Luise Abramowski.16
Miroslav Marcovich, however, is unimpressed, and proposes that ‘of the
apostles’ (tän ˆpost»lwn) should be added to the text at this point; this
would bring it into line with the phrase used in the next sentence (106.4):
‘in the memoirs of his apostles’, i.e. Christ’s apostles. An influential earlier
editor of Justin’s writings, J. C. Th. von Otto (1847), proposed a similar
emendment.
A decision is difficult, especially when we recall the parlous state of
the Parisinus codex. Justin’s repeated use of the phrase ‘memoirs of the
apostles’ does suggest that emendation may be appropriate. However, the
more difficult reading of the Parisinus codex is undoubtedly preferable. It is
the context which confirms that Justin is here referring to Peter’s memoirs,
i.e. Mark’s Gospel. In the same very complex sentence in which he refers
to the change of Peter’s name, Justin refers to the change of names of
the sons of Zebedee to ‘Boanerges, which is sons of thunder’ (106.3), a
phrase found at Mark 3.17, but not in the parallel passages in Matthew and
Luke.
There is a strong cumulative argument for Justin’s acceptance of at least
four gospels. Although the gospels are not named individually, there is
no doubt that Justin used Matthew’s and Luke’s Gospels extensively, and
Mark to a more limited extent. But what about John’s Gospel? This is a
controversial question, which can be referred to only briefly here.17 There

16 Abramowski, ‘Die “Erinnerungen”’, pp. 334–5.

17 For bibliography see J. W. Pryor, ‘Justin Martyr and the Fourth Gospel’, The Second Century 9 (1992)
153–69, esp. 153 n. 1.
102 Jesus and Gospel
is only one quotation to be considered, I Apol. 61.4-5: ‘Christ also said:
“Unless you are born again, you will not enter the kingdom of heaven”; for
it is clear to all that “it is impossible for those who have once been born to
enter into their mothers’ wombs”.’18 This is a free rendering of John 3.4-5,
not out of line with the way Jesus traditions are quoted elsewhere in Justin’s
writings.19 The phrase ‘kingdom of heaven’ (as in the similar tradition at
Matt. 18.3) is found in several manuscripts of John 3.3 (including the first
hand of Sinaiticus) as well as in numerous patristic witnesses.20 Justin may
have known the text of John 3.3 in this form, or he may have harmonized
John’s and Matthew’s phraseology, as others certainly did.
In addition to this free quotation, there are numerous allusions to pas-
sages in the Fourth Gospel, to say nothing of the probability that Justin
knew and developed the evangelist’s Logos doctrine.21 In 1943 J. N. Sanders
noted twenty-three possible allusions to John’s Gospel; others have com-
piled their own similar lists.22 There is a strong cumulative case, though this
raises two obvious problems. The more clearly one discerns the influence
of John’s Gospel, the more difficult it is to explain why Justin quotes it
only once. And, as we noted above, Justin refers to the sayings of Jesus as
‘short and concise’, hardly a natural way to refer to Jesus traditions in John’s
Gospel.
J. W. Pryor has recently argued that, while Justin knows the Fourth
Gospel, there is no evidence that he includes it among ‘the memoirs of
the apostles’. I do not think that we can rule out so firmly the possibility
that Justin’s reference to memoirs written by ‘apostles and those who fol-
lowed them’ includes the Fourth Gospel. I prefer to leave this as an open
question.

18 Kaª g‡r ¾ Crist¼v e«pen. ðAn mŸ ˆnagennhq¦te, oÉ mŸ e«s”lqhte e«v tŸn Basile©an tän oÉranän.
í Oti d• kaª ˆdÅnaton e«v t‡v mžtrav tän tekousän toÆv Œpax gennwm”nouv –mb¦nai, faner¼n
pŽs©n –sti.
19 See J. W. Pryor’s full discussion, with good bibliography, ‘Justin Martyr’, pp. 163–6. Pryor accepts that
I Apology 61.4-5 stems from John 3.3-5, but ‘it is not a case of direct borrowing by Justin himself, for
the saying of Jesus does bear the marks of having been changed under the influence of Matthew 18.3’
(p. 166). Osborn, Justin Martyr, p. 138, notes that Justin’s theology is not openly derived from the
Fourth Gospel. ‘The influence is shown on particular points and not on the shape of the whole.’
20 For details see esp. A. Huck and H. Greeven, eds., Synopse der drei ersten Evangelien (Tübingen:
Mohr, 1981), ad loc.
21 See M. J. Edwards, ‘Justin’s Logos and the Word of God’, JECS 3 (1995) 261–80. Edwards claims that
the roots of Justin’s Logos are in the Biblical tradition. Justin’s acquaintance with the Fourth Gospel
is left open.
22 J. N. Sanders, The Fourth Gospel in the Early Church (Cambridge: Cambridge University Press, 1943),
pp. 27–32. Osborn, Justin Martyr, p. 137, lists some twenty ‘coincidences of thought and expression’.
Pryor, ‘Justin Martyr’, pp. 158–9, adds some further examples of allusions to Sanders’s list, but notes
that some are more convincing than others.
 Justin Martyr and Irenaeus 103
While Justin does know a handful of traditions which did not find their
way into the canonical gospels,23 there is no evidence that he knew or
used an apocryphal gospel.24 Did he know, or compose, a harmony of seve-
ral gospels? Helmut Koester has recently claimed that the sayings Justin
included in his catechism were already harmonized in his Vorlage. In com-
posing this source, Justin or his ‘school’ did not intend to construct a
catechism, but was composing the one inclusive new gospel which would
make its predecessors, Matthew and Luke (and possibly Mark), obsolete.25
Although Koester’s theory has won some support,26 the limited evidence
he cites can be explained more plausibly along other lines.
It is much more likely that sayings of Jesus from the synoptic gospels
were harmonized for inclusion in the topically organized sets of sayings
mentioned above. While some of Justin’s harmonized traditions do seem
to have been used in his pupil Tatian’s more thorough-going harmony,27
there is no evidence to support the view that Justin intended to replace the
synoptic gospels. As we have seen, Justin’s own comments confirm that
he had a very high regard for gospels ‘written by the apostles and their
followers’ (Dialogue 103.8). So his preference for one single harmonized
gospel is inherently unlikely. There is no reason at all why Justin should not
have composed harmonized collections of sayings of Jesus for catechetical
purposes and have used them alongside written gospels. Indeed, in my view,
he almost certainly did just that.
Justin’s high regard for written gospels should by now be clear. Two
considerations provide further support. (a) In several passages Justin refers
to ‘reading’. I have already noted that Justin’s opponent Trypho is twice said
to have read appreciatively the sayings of Jesus ‘in the gospel’ (Dialogue 10.2;
18.1). In II Apology Justin’s opponent Crescens is accused of ‘running us
down without having read the teachings of Christ . . . or if he has read

23 The most notable are the references to the birth of Jesus in a cave (Dialogue 78.5); to the fire kindled
at the baptism of Jesus (Dialogue 88.3); and to an agraphon at Dialogue 47.5: ‘Our Lord Jesus Christ
said: “In whatsoever I overtake you, in that I will also judge you.”’ On the latter, see A. J. Bellinzoni,
‘The Source of the Agraphon in Justin’s Dialogue with Trypho 47.5’, Vig. Chr. 17 (1963) 65–70.
24 See especially Osborn, Justin Martyr, pp. 129–30; similarly, T. K. Heckel, Vom Evangelium des Markus
zum viergestaltigen Evangelium (Tübingen, Mohr, 1999), p. 326.
25 Helmut Koester, ‘The Text of the Synoptic Gospels in the Second Century’, in W. L. Petersen, ed.,
Gospel Traditions in the Second Century (Notre Dame and London: University of Notre Dame Press,
1989), pp. 28–33.
26 A. J. Bellinzoni, ‘The Gospel of Matthew in the Second Century’, SC 9 (1992) 197–258, esp. 239–42.
Miroslav Marcovich, Iustini Martyris Apologiae, p. 29, refers approvingly to Koester’s theory, but does
not discuss it.
27 See W. L. Petersen, ‘Textual Evidence of Tatian’s Dependence upon Justin’s APOMNHMONEY-
MATA’, NTS 36 (1990) 512–34.
104 Jesus and Gospel
them, he has not understood them’. In his own summary of Luke 24.25-6
and 44-6 and Acts 1.8-9 Justin notes that the risen Jesus ‘taught the disciples
to read the prophecies in which all these things were predicted as coming to
pass’ (I Apology 54.12, kaª ta±v profhte©aiv –ntuce±n).28 Luke’s narrative
implies oral teaching – indeed, it would have been difficult to read scrolls
while walking on the road to Emmaus. At this point, as in numerous other
passages, Justin’s narrative is very ‘bookish’.
(b) But what of the term ‘memoirs’ (ˆpomnhmoneÅmata)? Does this
square with the above observations concerning the ‘bookish’ character of
Justin’s writings? Although the term can refer to mere ‘notes’, it has now
been established by Niels Hyldahl that in Justin’s writings the term has
clear literary connotations. Hyldahl quotes Martin Dibelius approvingly:
‘An apologetic tendency is operative which is lifting up Christendom into
the region of culture. By means of the title “Memoirs” the Gospel books
would be classified as literature proper.’ Hyldahl notes that Socrates has
such a distinctive place in Justin’s writings that it was natural for him to
allude to Xenophon’s Memorabilia concerning Socrates in his choice of the
term ˆpomnhmoneÅmata.29
While it is easy to see why Justin would want to underline the literary
credentials of the gospels for apologetic purposes, we need not conclude that
he was exaggerating his case. I have recently argued that the Oxyrhynchus
papyri published in 1997 and 1998 suggest that, by the second half of the
second century, much earlier than has been usually assumed, the literary
qualities of the gospels and their authoritative status for the life and faith
of the church were widely recognized. The often-repeated claim that the
gospels were considered at first to be utilitarian handbooks written, by and
large, in a ‘reformed documentary’ style now needs to be modified, and
we need to remember that we do have a handful of second-century codices
with literary texts.30
For Justin, ‘the words of the Saviour’ were transmitted by the apostles in
written ‘upmarket’ memoirs which were known as gospels, though of course
he may well have known sayings of Jesus in other written or oral forms. A
close reading of all the evidence confirms the high regard in which Justin
held both the sayings of Jesus and the ‘memoirs of the apostles’. While it

28 Justin uses –ntugc†nw as ‘read’ in a number of passages; see, for example, I Apology 14.1; 26.8; 42.1;
44.12,13; 45.6; II Apology 3.6, 8; 15.3.
29 Niels Hyldahl, ‘Hesesipps Hypomnemata’, ST 14 (1960) 70–113. The quotation from Dibelius is
from his From Tradition to Gospels (London, Ivor Nicholson and E. Watson, tr. 1934), p. 40. See
especially Luise Abramowski’s full discussion, ‘Die “Erinnerungen”’.
30 See Chapter 9 in this volume.
 Justin Martyr and Irenaeus 105
is true that Justin does not refer explicitly either to the sayings or to the
memoirs as ‘Scripture’, he comes within a whisker of doing so. Like the
‘Scriptural’ prophets, the ‘memoirs’ are read at length and expounded in
the liturgical Sunday gatherings of Christians.

4.2 irenaeus
Irenaeus wrote his Adversus Haereses about ad 180, barely a generation
after Justin composed his Apologies and Dialogue. Irenaeus knows Justin’s
writings,31 and may even have met him in Rome. Although both writers
hold the gospels and especially the words of Jesus in high regard, there are
important differences. Whereas Justin made limited use of John’s Gospel,
Irenaeus has no hesitation in accepting its authority. Indeed, it is arguable
that this gospel influenced Irenaeus’ theological thought more deeply than
any other writing.
Justin seems to have known four gospels, though he never names any
of the evangelists and in only one passage does he show even the slightest
interest in the plurality of the gospels.32 Irenaeus is more specific: two
gospels were written by named apostles or disciples (Matthew and John)
and two by their followers (Mark and Luke) (Adv. Haer. iii.10.1; 10.6; 11.1).
His lengthy and sophisticated defence of the fourfold Gospel takes us far
beyond Justin. His line of argument strongly suggests that he is not making
a case for a recent innovation, but underpinning what he and others had
long accepted, i.e. that the church had been given one Gospel in fourfold
form – four authoritative writings, no more, no less.
Both writers give the same status to the gospels as they do to the OT
writings. Although Justin is familiar with the term grafž for Scripture
(e.g. Dialogue 56.12,17), he never uses this term to refer to a gospel, nor
does he himself use ‘it is written’ (g”graptai) to introduce either an OT
quotation or a saying of Jesus as Scripture. Irenaeus, however, does refer
to the gospels explicitly as ‘Scripture’: ‘since all the Scriptures, both the
prophets and the Gospels proclaim this’ (‘cum itaque universae Scripturae,
et prophetiae et Euangelii . . . praedicent’, Adv Haer. ii.27.2). At the end of
Book ii, he announces that in his next book he will support his argument
from ‘divine Scripture’; in Book iii there are very many more references

31 The extent of this knowledge merits further investigation. For discussion of the similar use of
Matt. 7.15 by Justin and Irenaeus, see D. J. Bingham, Irenaeus’s Use of Matthew’s Gospel in Adversus
Haereses (Leuven: Peeters, 1998), pp. 27–32.
32 Dialogue 103.8. See above, pp. 100–01.
106 Jesus and Gospel
to the gospels than to the OT, so there is a clear implication that they are
‘Scripture’.
Nonetheless, Irenaeus remains somewhat coy about referring to the
gospels as Scripture. Only once does he introduce a saying of Jesus with
‘Scripture says’, and even in this case Matt. 13.18 is alluded to rather than
quoted (iv.41.2). In a handful of places he introduces a verse from the
gospels with ‘it is written’; see, for example, ii.22.3; iv.20.6. For Irenaeus
‘Scripture’ is first and foremost the OT, though it is quite clear that the
gospels and sayings of Jesus enjoy the same level of authority as the OT.33
We noted above that Justin is adamant that traditions of the sayings and
actions of Jesus have been transmitted carefully in the written gospels by
the apostles, and are handed on and carried out by his fellow Christians. In
the introductory sections of the First Apology Justin emphasizes the careful
transmission of tradition by using repeatedly the verb parad©dwmi. The
same points are made even more strongly by Irenaeus. In his introduction
to Book iii, where he is concerned above all with the status of the Gospel
in fourfold form, he stresses that the oral proclamation of the Gospel by
the apostles was ‘later handed down to us in the Scriptures’ (iii.1.1, ‘in
Scripturis nobis tradiderunt’; –n grafa±v par”dwkan ¡m±n).34 Irenaeus
then refers briefly to the origin and authorship of the individual gospels.
His comment on Mark, the disciple and interpreter of Peter, echoes the
phraseology just quoted: ‘he handed down to us in written form the things
proclaimed by Peter’ (‘ipse quae a Petro adnuntiata per scripta nobis tradidit’
(iii.1.1).35
At the close of his extended comments on the origin and authority of
the gospels, Irenaeus summarizes his key points. The Gospel has been
transmitted in written form by the apostles; since God made all things in
due proportion and adaptation, it was fit also that the outward aspect of
the Gospel should be well arranged and harmonized (iii.11.12, ‘oportebat
et speciem Euangelii bene compositam et bene compaginatam esse’). In
this closing section Irenaeus states three times over that the gospels can be
neither more nor fewer in number than they are (iii.11.8.1, ‘neque autem
plura numero quam haec sunt neque rursus pauciora capit esse Euangelia’;
cf. 11.9.1 and 11.9.12, where almost identical terminology is used).

33 For these comments on Irenaeus’ use of ‘Scripture’ I am indebted to Benoit, Saint Irenée, pp. 120–2.
34 I have cited the Latin text and Greek retroversion from Sources chrétiennes 211, ed. A. Rousseau and
L. Doutreleau (Paris: Cerf, 1974).
35 At this point we have Eusebius’ version of the original Greek: kaª aÉt¼v t‡ Ëp¼ P”trou khruss»-
mena –ggr†fwv ¡m±n parad”dwken (HE v.8.3).
 Justin Martyr and Irenaeus 107
‘No more, no less’ almost becomes a slogan for Irenaeus. His terminology
is so closely related to a ‘canon formula’ widely known in antiquity that it
is most surprising that Irenaeus does not draw on it.36 As W. C. van Unnik
noted, the ‘canon formula’, ‘neither add nor take away’, has deep roots
in both Biblical and Greek thought.37 Irenaeus knows the concepts in the
same order, ‘no more, no less’.
From Aristotle onwards, a ‘canonical’ work was defined as one to which
nothing could be added and from which nothing could be subtracted
without harming its aesthetic unity. Aristotle states that ‘neither add nor take
away’ is a proverbial expression.38 The ‘canon formula’ was a well-known
slogan in the Hellenistic world in the realm of aesthetics. Irenaeus places
a great deal of weight on the aesthetic unity of the fourfold Gospel, so
that his failure to quote the slogan is baffling, especially as Eusebius refers
to it.39
Justin emphasizes the importance and power of the words of Jesus. As
we noted above, several passages confirm that he is concerned primarily
with their written form rather than with continuing oral tradition. Justin
does not state that ‘the words of the Saviour’ are available anywhere but
in ‘the memoirs of the apostles’. I am convinced that this is also the case
with Irenaeus, though here I am somewhat out of line with the current
consensus. Hans von Campenhausen claims that Irenaeus does not think
of the gospels as sources for the words of Jesus. Their purpose is simply to
provide documentary evidence of the teaching of ‘that apostle’ who wrote
down the gospel; the words of the Lord are treated on their own, without
reference to the gospels.40 Y.-M. Blanchard has recently gone much further
down this path: ‘ainsi, au temps d’Irénée, la mémoire vivante des logia du
Seigneur paraı̂t constituer le canal privilégié de la Tradition chrétienne’.41
According to Blanchard, the gospels are of secondary importance; pride of
place among the four is given to Luke.42

36 See, however, iv.33.8: ‘neque additamentum, neque ablationem recipiens’.

37 W. C. van Unnik, ‘De la règle Mžte prosqe±nai mžte ˆfele±n’, Vig. Chr. 3 (1949) 1–36. See also
Christoph Dohmen and Manfred Oeming, Biblischer Kanon, warum und wozu?: eine Kanontheologie
(Freiburg: Herder, 1992), especially pp. 78–89; John Barton, The Spirit and the Letter: Studies in the
Biblical Canon (London: SPCK, 1997), pp. 133–4.
38 Aristotle, EN ii.1106b.
39 At HE v.16.3 Eusebius cites an anonymous letter which refers to the formula with reference to the
‘canon’ of the NT.
40 Von Campenhausen, Formation, pp. 191 and 202. John Barton summarizes approvingly von Camp-
enhausen’s position in his The Spirit and the Letter, pp. 82–4.
41 Y.-M. Blanchard, Aux sources du canon, le Témoinage d’Irenée, Cogitatio Fidei 174 (Paris: Cerf, 1993),
p. 221.
42 Ibid., pp. 206 and 229.
108 Jesus and Gospel
Irenaeus does attach special importance to the sayings of Jesus, but it is a
Christological, not a hermeneutical, priority.43 Irenaeus does not single out
the words of Jesus as a ‘canon within the canon’. He sees the gospels as the
records of the teaching of Jesus. This is nowhere clearer than in the Preface
to Book iii: ‘The Lord of all gave to his apostles the power to proclaim the
Gospel, and from them we have known the truth, that is to say the teaching
of the Son of God. And it is to them that the Lord said: “Anyone who hears
you hears me.” Irenaeus devotes Book iv to the ‘words of the Lord’, but
at iv.6.1 there is a further clear indication that they are not to be thought
of in isolation from their written form in the gospels.44 Irenaeus cites
Matt. 11.27 = Luke 10.22, and then comments: ‘Thus has Matthew set it
down, and Luke similarly, and also Mark; for John omits this passage.’ He
is, of course, mistaken about Mark, but that is beside the point. In several
passages in Book iv the narrative context of the cited ‘words of the Lord’
in one of the gospels (usually Matthew) is retained; in other passages the
deeds of Jesus are referred to or summarized.45
It would be rash to claim that Irenaeus does not know oral traditions
of the sayings of Jesus. But his vigorous exposition of the fourfold form of
the Gospel leaves little room for continuing oral tradition. If one were to
ask either Justin or Irenaeus where one could find the sayings of Jesus, the
answer would surely be, ‘in the writings of the apostles and their followers’.

Justin and Irenaeus both hold the sayings of Jesus and the gospels in high
regard, and on a level with the OT Scriptures. In this respect the simi-
larities between the two second-century giants are as important as the
differences. Although Irenaeus takes several more steps towards acceptance
of a ‘canon’ of four written gospels than does Justin, the great teacher of the
new ‘philosophy of Christ’ paves the way. I have emphasized more strongly
than most the importance of written Jesus traditions for both Justin and
Irenaeus. There is an obvious corollary: we may have allowed Papias’ prefe-
rence for ‘the living voice’ over ‘the written word’ to influence too strongly
our reading of both Justin and Irenaeus.
The physical appearance of early Christian writings at the time in ques-
tion is regularly overlooked in discussions of their status. I have argued
elsewhere that the emergence of the four-gospel canon is related to the
dissemination of the four gospels in codex form. Justin may well have had
43 See especially the Preface to Book iv, and iv.1.1. See also Bingham, Irenaeus’s Use of Matthew’s Gospel
in Adversus Haereses, pp. 97–8.
44 Not surprisingly, Blanchard, Aux sources du canon, fails to discuss this passage.
45 For the former, see, for example, iv.12.4-5; 10.1; 29.1. For the latter: iv.8.2.
 Justin Martyr and Irenaeus 109
a four-gospel codex in his catechetical school in Rome by about ad 150.
As long ago as 1933, F. G. Kenyon suggested that Irenaeus may have been
accustomed to the sight of codices which contained all four gospels. The
evidence for this conclusion is now much stronger than it was seventy years
ago.46 Today, when we hear vociferous claims on behalf of the Gospels of
Peter and Thomas, we need to recall that there is no manuscript evidence
for the acceptance of any ‘fifth’ gospel alongside one or more of the writ-
ings of the fourfold Gospel. Codex and canon go hand in hand, but that
is another story.
In discussions of the emergence of the canon, whether of the OT or the
NT writings, definitions are all important, and the devil is in the detail.
Even though Irenaeus does not use the term ‘canon’ in its now customary
sense, his insistence that the one Gospel proclaimed by the apostles is found
in four written gospels, no more, no less, implies a ‘closed’ gospel canon.
If our definition of ‘gospel canon’ includes reference to an agreed list of
widely accepted authoritative writings, then it did not exist at the end of the
second century. Irenaeus was a towering figure, but we must not assume that
his views on the fourfold Gospel were accepted universally. Justin Martyr
influenced Irenaeus strongly, but Justin’s pupil Tatian, whose Diatessaron
almost won the day, took a very different path.47

46 See Chapter 8, pp. 165–7 below.

47 After I had completed this chapter my attention was drawn to the following: W. A. Lohr, ‘Kanons-
geschichtliche Beobachtungen zum Verhältnis von mundlicher und schriftlicher Tradition im zweiten
Jahrhundert’, ZNW 85 (1995) 234–58.
 chap t e r 5

The law of Christ and the Gospel

There are some baffling phrases hidden in early Christian writings which
are worth careful examination.1 The phrase ‘the law of Christ’ is one such.
Although it is used only once in the New Testament (Gal. 6.2), it teases
exegetes, it raises central questions of theological method, and it still forces
us to ask awkward questions.2 Was this phrase part of Paul’s Gospel? And
should it be part of Christian proclamation today?
In his influential commentary, H. D. Betz insists that Gal. 6.2, and
indeed all the ethical directives in Gal. 5.13-6.10, are not directly derived
from the Gospel that Paul preached.3 Richard Hays has argued, surely
correctly, that this disjunction of kerygma from conduct arises from an over-
emphasis on individualistic soteriological elements at the expense of the
corporate dimension in Paul’s theological thought.4 Paul’s encouragement
to the Galatian Christians to ‘fulfil the law of Christ’ (Gal. 6.2) was surely
part of the Gospel message he wished to convey to the Galatian churches
and no mere ethical addendum.
But that leaves us with a further set of questions. Why did Paul not
make good theological capital out of the phrase in his other letters? Did
the apostle decide that the phrase was too ambiguous or too prone to
misunderstanding to merit further use? If so, should we follow his lead and
drop it from contemporary theological reflection and from liturgies? Is this
a phrase which has ‘punched above its weight’ for far too long?

1 An earlier version of this chapter was given as a paper in a series of special seminars in Cambridge
which formed part of the celebrations in 2002 of the 500th anniversary of the establishment of the
Lady Margaret’s Professorship. It was published in D. F. Ford and G. N. Stanton, Scripture and
Theology: Reading Texts, Seeking Wisdom (London: SCM, 2003), pp. 169–84.
2 See especially R. B. Hays, ‘Christology and Ethics in Galatians: The Law of Christ’, CBQ 49 (1987)
268–90; M. Winger, ‘The Law of Christ’, NTS 46 (2000) 537–46. Winger claims that the phrase ‘the
law of Christ’ is more likely to mislead than to instruct. Paul ‘probably thought it clearer than it has
turned out to be; every writer has that experience’ (p. 545).
3 H. D. Betz, Galatians, Hermeneia Commentary (Philadelphia: Fortress, 1979), p. 292.
4 Hays, ‘Christology and Ethics’, pp. 270–2.

110
 The law of Christ and the Gospel 111
On the other hand, we may decide that statistics can mislead. Paul may
have used the phrase only once, but we should not forget that he also
referred to the Lord’s Supper in only one of his letters. As we shall see, there
is a handful of related phrases and themes in Paul’s letters and elsewhere in
the New Testament. And in several early second-century writings similar
phrases are prominent. Although the phrase often suffered benign neglect
in later centuries,5 from time to time this gem has been dusted down and
polished, and sometimes partly recut.
‘The law of Christ’ has played at least one major role on the stage of
history. John Fisher, the first holder of the Lady Margaret’s Professorship
of Divinity at Cambridge, used the phrase at a turning point in his dispute
with King Henry VIII. When Henry began to seek a divorce from Catherine
of Aragon in 1527, Fisher soon became the Queen’s foremost defender. In
1531 he wrote the mitigating words by which the clergy in Convocation
qualified their earlier acceptance of the King as their Supreme Head on
earth. The clergy acknowledged the King only ‘as far as the law of Christ
allows’ (‘quantum per legem Christi licet’). This move set in motion a train
of events, one of which was the beheading of John Fisher in 1535. Why
did Fisher appeal to ‘the law of Christ’ at such a tense time? What did he
understand by this phrase?
The phrase had not been part of Thomas Aquinas’ vocabulary; he pre-
ferred ‘the new law’. And, although John Fisher had close contact with
Erasmus, he doesn’t seem to have been influenced by Erasmus in his choice
of this phrase. For in none of the five editions of the Annotationes does
Erasmus say a single word about ‘the law of Christ’ in his comments on
Gal. 6.2.6 Perhaps Fisher chose the phrase because it was vague and ill-
defined, but sounded good.7 Or perhaps Fisher picked up the phrase quite
deliberately from Gal. 6.2. If the latter, then I would love to know how he
interpreted Paul’s baffling words!
Martin Luther commented powerfully on the phrase in his 1519 and 1535
commentaries on Gal. 6.2, but it played no more than a very minor role
in his expositions of the two uses of the law or in the later controversies
5 Bernard Häring’s The Law of Christ (3 vols., Cork: Mercier, 1963) is only a partial exception. In spite
of its title, the phrase ‘the law of Christ’ is not discussed in this very traditional Roman Catholic
moral theology until pp. 252–63. Häring seems to take the phrase to refer to the ethical teaching of
the NT as a whole and to be synonymous with Thomas Aquinas’ frequently used phrase, ‘the new
law’. I am grateful to my colleague Dr Markus Bockmuehl for this reference and for a number of
helpful comments on an earlier draft of this chapter.
6 See Anne Reeve, ed., Erasmus’s Annotations on the NT. Galatians to the Apocalypse: Facsimile of the
Final Latin Text with All Earlier Variants (Leiden: Brill, 1993).
7 So my colleague Dr Richard Rex in a written communication dated 30 January 2002. See his The
Theology of John Fisher (Cambridge: Cambridge University Press, 1991).
112 Jesus and Gospel
over the ‘tertium usus legis’. Neither John Calvin nor Karl Barth found any
theological use for the phrase. And, since very few contemporary systematic
theologians bother to wrestle with exegesis, we shall be surprised to find
much interest in the phrase in their writings.
As we shall see, over the centuries, whenever ‘the law of Christ’ has found
favour, it has been understood in several different ways. Which of those
interpretations, if any, can still play a part in the Christian Gospel today?
I shall start with Paul’s use of the phrase at Gal. 6.2. What was Paul’s
intention when he first used the phrase? Should his understanding deter-
mine any continuing theological use today? I shall then turn to this phrase’s
closest relatives elsewhere in the NT and in early Christian writings. Do
the second-century uses of the phrase help us theologically? ‘The law of
Christ’ raises perennial questions for Biblical scholars and for theologians,
and for those who want to wear both hats.
In this chapter I hope to show that there are themes associated with this
phrase which can still be put to good use as part of the Christian Gospel.
We may continue to use the well-known intercessory prayer based on
Gal. 6.2: ‘Help us so to bear the burdens of others that we may fulfil
the law of Christ.’ But if we do so, we shall need to be clear what we are
doing when we attempt to bear the burdens of others.

5.1 fulfilling the l aw of christ : gal. 6.2

Here is the thirty-first of Paul’s references in Galatians to n»mov, law. In
nearly all the preceding instances the reference is to the law, i.e. the law
of Moses.8 So it is most unlikely that, without alerting his listeners, Paul
changes tack and refers to the teaching of Jesus as ‘law’, or to ‘showing love
for others’ (i.e. bearing the burdens of others) as ‘law’.
Ever since Gal. 2.16, ‘law’ has been pounding relentlessly in the Galatians’
ears with a negative beat. However, at 4.21b, 5.14, and 6.2 Paul’s tone
changes dramatically: n»mov is used in a positive sense in all three verses.
The reference at 5.14 to fulfilling the law in loving one’s neighbour is
particularly striking; it paves the way for 6.2, for in both verses verbs from
the same root are used. Although ‘Christ’ and ‘law’ have regularly stood in
stark contrast earlier in the letter, at 6.2 they are brought together in an ear-
catching but baffling phrase. Since ‘fulfilling the law’ at 5.14 clearly refers
to the law of Moses, the use of the similar verb at 6.2 strongly suggests that
‘law’ here also refers to the law of Moses, and not to a norm or principle.

8 J. L. Martyn, Galatians, Anchor Bible Commentary (New York: Doubleday, 1997), p. 555.
 The law of Christ and the Gospel 113
‘Carry one another’s burdens’, urges Paul, ‘and in this way you will fulfil
the law of Christ.’9 The careful listener to this letter being read aloud in
the Galatian churches can hardly have failed to miss Paul’s insistence that
Christ himself ‘bore the burdens of others’ and so provided an example
for the Galatian Christians. Paul’s expansions of his opening ‘grace and
peace’ formula in his letters often foreshadow several of the letters’ main
theological themes. This is certainly the case in Galatians. At Gal. 1.4 Paul
notes that Christ ‘gave himself for our sins’. Christ’s self-giving love forms
the climax of the richest section of the letter, Gal. 2.15-20: ‘the Son of God
loved me and gave himself for me’. Christ has fulfilled the law himself in
his self-giving in love for others.10
So in Galatians we have one answer to the question: what is the law of
Christ? In Gal. 6.2 it is the law of Moses interpreted by Christ, with the ‘love
commandment’ and ‘carrying the burdens of others’ as its essence; it is fulfilled
by Christ in his own self-giving love.11
Several further comments are in order. (1) With their positive references to
law, Gal. 4.21b, 5.14, and 6.2 stand in counterpoise to Paul’s other references
to the law of Moses in Galatians. Taken together, these verses confirm that,
in spite of the numerous negative comments on the law elsewhere in this
letter, Paul did not repudiate the law of Moses, as some of his later followers
(most notably Marcion) and some of his opponents (see Acts 21.28) wrongly
supposed.
(2) The immediate context emphasizes that those who live by the Spirit
are not free to gratify the desires of the flesh (5.16).12 However, Paul does
not spell out the precise ways in which the law of Moses is to be retained
now that believers in Christ have been set free from the present evil age
(1.4). No one will be right-wised by God on the basis of carrying out the
law of Moses (2.16), but the law is not to be ignored or discarded, for it is
not opposed to the promises of God (3.21).

9 Dr Michael Thompson has drawn my attention to the partial parallel at Rom. 15.1-3, where encour-
agement to bear (bast†zein) the failings of the weak is linked to the example of Christ.
10 Winger, ‘The Law of Christ’, p. 538, protests that, in order to find a reference to the example of the
self-giving love of Christ at Gal. 6.2, Hays must reach ‘back all the way to 2.20’. But he fails to note
the importance of 1.4, and sweeps aside 3.13 and 4.4-5 too readily.
11 Cf. John Barclay’s explanation, Obeying the Truth (Edinburgh: T. & T. Clark, 1988), p. 134: the law of
Christ is ‘the Law as it has been taken in hand by Christ himself ’; and also J. L. Martyn’s comment
on the phrase (Galatians, ad loc.): ‘the law as it has fallen into the hands of Christ’.
12 Winger, ‘The Law of Christ’, helpfully emphasizes the importance of Gal. 5.16-25 for the interpreta-
tion of 6.2. He claims that, for Paul, ‘those who live according to the Spirit follow the law of Christ –
or better, as Paul says, they bring it to realization’ (p. 544). However, I am not convinced that Paul
uses ‘law’ at Gal. 6.2 ‘in a somewhat looser sense, not as identifying any specific, legal instruction,
but as referring to the way Christ exercises his lordship over those called by him’.
114 Jesus and Gospel
Paul continued to mull over this antinomy. The apostle takes a further
step in Rom. 13. 8-10. ‘Love is the fulfilling of the law’, to be sure, but this
axiom does not mean that the commandments concerning adultery, mur-
der, theft, and covetousness may be ignored. This passage seems to have
encouraged the later strong Christian conviction that the ten command-
ments are the core of the Mosaic law; they (and, for some, they alone) have
abiding significance for Christians.
(3) The encouragement to ‘bear one another’s burdens’ is almost certainly
taken from the Socratic tradition and the Greek doctrines about ‘friend-
ship’.13 A well-known maxim is transposed by Paul into a new key. We
should welcome Paul’s appropriation of the conventional ethical teaching
of his day at several points in Galatians 5 and 6, and especially in the lists
of virtues and vices at Gal. 5.19-23. For, in so doing, Paul has given us an
example to follow. Why should we not appropriate some of the insights
of the moral philosophers of our day and set them in a firmly constructed
theological framework, as Paul himself does?
I do not think that Paul is picking up a slogan used by the agitators in
Galatia and throwing it back in their face. Nor do I think that Paul is being
merely playful, as has been suggested. In the preceding argument of the
letter Paul has carefully prepared his listeners and readers for his dramatic
use of the phrase at 6.2. His failure to spell out its meaning more fully may
indicate that the Galatian house churches had already been well drilled
by Paul in its meaning. This is an interesting possibility, which has been
overlooked in the voluminous literature. In his letters Paul does use a series
of key ‘short-hand’ phrases whose content his recipients could readily fill
out for themselves.14 ‘The law of Christ’ may be one such phrase.
Not long after writing Galatians, Paul wrote I Corinthians. In a highly
rhetorical paragraph in chapter 9 Paul comments on his missionary strat-
egy. In v.20 he explains that when preaching to Jews he was prepared to
exercise a measure of self-renunciation. In v.21 he comments on his strategy
when preaching to Gentiles: ‘I identified as one outside Mosaic jurisdiction
with those outside it.’ A rider is added immediately: ‘Of course, I am not
outside God’s law (Šnomov qeoÓ), but I am ›nnomov CristoÓ.’ The latter
phrase is often translated, ‘I am under the law of Christ’, which is adequate,
though it suggests that the Greek here is the same as in Gal. 6.2. Given the
subtle play on words in I Cor. 9.20 and 21, it was hardly an option for Paul
13 Betz, Galatians, p. 549.
14 See especially Margaret Mitchell, ‘Rhetorical Shorthand in Pauline Argumentation: the Functions
of “the Gospel” in the Corinthian Correspondence’, in L. A. Jervis and P. Richardson, eds., Gospel
in Paul, FS R. N. Longenecker (Sheffield: JSOT Press, 1994), pp. 63–88. See also pp. 49–52 above.
 The law of Christ and the Gospel 115
to repeat the phrase used in Galatians, ¾ n»mov toÓ CristoÓ. In fact, as
far as we know, the apostle never reused that phrase.
What did Paul mean by ›nnomov CristoÓ? ‘I am under Christ’s juris-
diction’ catches the sense. The preceding chapters of I Corinthians suggest
that the phrase may include commitment to sayings of Jesus, but to limit
it in that way would miss its Christological thrust.
Paul’s use of the phrase ‘the law of faith’ at Rom. 3.27 has teased many
a commentator. Although many claim that Paul here uses n»mov to mean
‘principle’ or ‘rule’, it is most unlikely that Paul jumps without warning,
from using n»mov to refer to the law of Moses, to the more general sense of
‘principle’, and in any case the context will not allow this reading. We must
take our cue from v.31, the final step in Paul’s argument. ‘Do we overthrow
the Torah on the basis of faith in Jesus?’ By no means: we uphold the law.
So ‘the law of faith’ is the law discerned and obeyed on the basis of faith in
Jesus. In other words, this phrase is a first cousin of ¾ n»mov toÓ CristoÓ.15

5. 2 sim il ar phrases outsid e the pauline corpus

At Jas. 1.25, 2.8, and 12 the writer qualifies n»mov three times. His other uses
of n»mov refer to the law of Moses or to one of its specific commandments.
So it is probable that ‘the perfect law of freedom’, ‘the royal law’, and ‘the law
of liberty’ are almost synonymous references to the law understood from a
Christian perspective; in the context of this letter, that means the teaching
of Jesus.16 In other words, we are not far from ‘the law of Christ’. This is
especially clear at Jas. 2.8: ‘You are doing well if you are carrying out the
royal law according to the Scripture, “Love your neighbour as yourself.”’
There is a clear implication that the love commandment lies at the heart
of the law (cf. Matt. 22.34-40).
What is implicit in James is much more explicit in Matthew’s Gospel. The
evangelist has taken great care over the composition of his five discourses
because he values the sayings of Jesus so highly. The sayings of Jesus are to
be prominent in the missionary proclamation and catechetical instruction
of the ‘new people’ (28.18-20). The closing verses of the Sermon on the
Mount emphasize strongly the importance of hearing and acting on the

15 Perhaps Rom. 8.2 is also a first cousin of Gal. 6.2: in Christ Jesus the law (of Moses) becomes
life-giving (cf. 7.10 and 14a) and of the Spirit.
16 So also Sophie Laws, The Epistle of James, Black’s NT Commentaries (London: A. & C. Black, 1980),
p. 110: ‘It is probable that when James quotes Lev 19.18 as scripture he does so in the knowledge that
this scripture has received the added authority of Jesus’ use.’
116 Jesus and Gospel
words of Jesus (7.24-7). For Matthew ‘the will of the heavenly Father’ is
equated with carrying out the sayings of Jesus (7.21, cf. Luke 6.46).
Matthew’s Gospel provided the ‘new people’ with a set of authoritative
traditions to be set alongside the law and the prophets. The evangelist does
not spell out as clearly as his modern interpreters would like the precise
relationship between ‘new’ and ‘old’. Matthew’s Jesus does not repudiate
the law: its continuing importance is affirmed very strongly (5.17-19). The
love commandment is singled out by Jesus as expressing the very essence of
Scripture (7.12; 22.37-9), but in no way does this contradict the law, any
more than do the so-called antitheses in 5.21-48. Matthew hints – but no
more – that the sayings of Jesus are the criterion for the interpretation of the
law, but his primary emphasis is on the ways the sayings of Jesus strengthen
and fulfil the law and the prophets.
In Matthew, as elsewhere in the NT, Christology and ethics are linked
inextricably. The full significance of the teaching of Jesus can only be
discerned in the light of the conviction that Jesus is ‘God with us’
(Matt. 1.23). With his coming, in fulfilment of Isa. 9.1, ‘light has dawned’
(4.16). The Sermon on the Mount is proclamation of the good news of
God’s kingly rule (4.17, 22). It is both gift and demand; it is not a new set
of rules and regulations.
Matthew does not use the phrase ‘the law of Christ’, but he would not
have been unhappy with it. For Matthew, the person and words of Jesus are
the criterion by which the law is to be interpreted. Although some exegetes
argue that the teaching of Jesus is also in view at Gal. 6.2, in my opinion
they are mistaken. However, in spite of differences, Matthew and Paul agree
that ‘the law of Christ’ has as its focal point the love commandment and
has a Christological reference. Both notions were to be important in later
theological exposition of our phrase.
The writer of I John insists at 2.3 that we know God if we keep God’s
commandments. The phraseology recalls Sirach 29.1 and Matt. 19.17, and
we are only one step from Tobit 14.9, Acts 15.5, and James 2.10, where the
verb thre±n is used of keeping the law of Moses. Three verses later the
writer’s train of thought turns to Christ: whoever says, ‘I abide in him’
ought to walk just as Christ walked (I John 2.6). He then refers to the
old commandment which his readers have had from the beginning, the
‘word’ that they have heard. What is it? It is of course Christ’s command,
‘love one another’. In short, the writer implies that the love commandment
is the focal point of God’s commandments, i.e. the law of Moses. There
is certainly no suggestion that the commandment to love is not found
 The law of Christ and the Gospel 117
in the OT, for at 3.11-12 Cain is attacked for failing to love his brother
Abel.
Although n»mov is not used in the Johannine letters, we are not far from
the thought that ‘the law of Christ’, and in particular the commandment to
love one another, is the lens through which God’s commandments should
be read and obeyed.
A similar pattern of thought is found in II John 6-10. God’s command-
ments and Christ’s command to love one another are juxtaposed and all
but fused together. But this letter adds a note not found explicitly in I John:
Christ’s teaching is referred to twice over (verses 9 and 10), and it is fairly
clear that this teaching extends beyond the love commandment.
The preceding paragraphs have confirmed that there is a strikingly simi-
lar pattern to be observed in several NT writings – a generally overlooked
show of unity amidst diversity. There are almost certainly no direct lite-
rary relationships between Paul, James, Matthew, and the writers of the
Johannine letters. Yet in all four strands of earliest Christianity we find a
continuing commitment to the law of Moses alongside an insistence that it
should be understood from a new perspective. That new perspective is not
sketched out fully, but in every case it includes a Christological element,
and either the sayings of Jesus or the love commandment, or both.
This theological pattern can be traced down through the centuries when-
ever our theological gem ‘the law of Christ’ is dusted down. Sometimes
Christology is more prominent, sometimes the teaching of Jesus, some-
times the love commandment. I shall now give the merest sketch to back
up that contention.

5.3 second -century writ ers

In the early decades of the second century the Shepherd of Hermas takes
an interesting step beyond NT writers. The Son of God is the lawgiver:
he has given to the people the law given to him by the Father (Sim. 5.6, 3;
ch. 59). The Son of God is himself ‘the law of God’ given to the whole
world (Sim. 8.3, 2-3; ch. 69).
Christ is also called ‘the law’ three times in the Kerygma Petri, which
dates from the same period. Alas, all we have from this treatise is a handful
of quotations in later writers, especially Clement of Alexandria. Christians
worship God in the manner neither of the Greeks nor of the Jews. ‘We
worship God through Christ in a new way.’ ‘For we have found in the
Scriptures, how the Lord says: “Behold, I make with you a new covenant,
not as I made (one) with your fathers in Mt Horeb.” A new one he made
118 Jesus and Gospel
with us. For what has reference to the Greeks and Jews is old. But we are
Christians, who as a third race worship him in a new way.’17 Here we have
early evidence for a very radical sense of self-understanding on the part of
Christians. Scripture is not abandoned, but Christ himself has become ‘the
law’.
A fuller exposition of the same theme is found in Justin Martyr’s Dialogue
with Trypho, written c. ad 160, possibly in dependence on the Kerygma
Petri.18 For Justin, Christ is himself given by God as the final and eternal
law (Dialogue 11.2, twice), and the new law (11.4, ¾ kain¼v n»mov).19 Justin
elaborates his point in a striking credal passage in 43.1: by the will of
the Father, Christ was born Son of God by means of the Virgin; he was
proclaimed by the prophets as about to come ‘as an everlasting law and
a new covenant for the whole world’. Justin insists that the law given by
God to Moses at Mt Horeb is antiquated and belongs to Trypho and his
fellow-Jews alone, whereas Christ, the new law, was given to all people
(11.2). Justin does not believe for one moment that the law of Moses has
been abolished, though he does have difficulty spelling out precisely how
Christians should use it, as, of course, did Paul himself.
Justin also understands ‘the law of Christ’ in a different sense. His Jewish
opponent Trypho issues a strong challenge: Justin and his fellow-Christians
claim to worship God, but fail to make their lives different from those of
Gentiles in that they keep neither the feasts nor the sabbaths, nor do they
have circumcision, nor do they carry out God’s commandments (10.3). In
short, Christians are law-less. As a key part of his extended response, Justin
refers to Christ as the new lawgiver (¾ kain¼v nomoq”thv, 12.2 and 14.3),
through whom ‘the poor have the Gospel preached to them, and the blind
receive their sight’ (cf. Matt. 11.6 = Luke 7.22). The sayings of Jesus are
understood here as the gift of the new lawgiver. For Justin the teaching of
Jesus embedded in the gospels is ‘the law of Christ’.20
So, for Hermas and for Justin, Christ is himself both the new law and
the new lawgiver, but without the implication that the law of Moses is
to be abandoned. The Epistle of Barnabas comes perilously close to the
17 For the texts and bibliography, see J. K. Elliott, The Apocryphal New Testament (Oxford: Clarendon,
1993), pp. 20–3; more fully, W. Schneemelcher, ed., New Testament Apocrypha, Vol. ii (Cambridge:
Clarke, 1992), pp. 34–41.
18 So O. Skarsaune, The Proof from Prophecy. A Study in Justin Martyr’s Proof-Text Tradition: Text-Type,
Provenance, Theological Profile (Leiden: Brill, 1987), pp. 72–3.
19 The most accessible translation is by A. Lukyn Williams, Justin Martyr: The Dialogue with Trypho
(London: SPCK, 1930).
20 For a fuller discussion, see Chapter 4. Although Justin does not refer explicitly to Paul or to his
letters, at numerous points he betrays knowledge of them. So it is possible that his understanding
of Christ as the new lawgiver has been influenced by Gal. 6.2.
 The law of Christ and the Gospel 119
latter notion: the Lord has abolished sacrifices, new moons, and sabbaths
so that ‘the new law of our Lord Jesus Christ which is free from the yoke
of compulsion might have its offering, one not made by man’ (2.6). For
Barnabas the Mosaic law has mistakenly been interpreted literally rather
than spiritually as intended (10.2, 9).

5.4 the t eaching of jesus as ‘the l aw of christ ’

Justin Martyr’s notion of Christ as the ‘new lawgiver’ became all pervasive
in the later tradition of the church. I shall now refer briefly to a handful of
examples. First of all, the magnificent thirteenth-century statue of Christ the
lawgiver from the south side of Chartres Cathedral in France. Here Christ
is depicted holding up one arm in blessing. His other hand is offering a
beautifully bound codex, which I take to be the gospels. In other words,
Christ is offering his own teaching as a ‘new law’. Protestants underestimate
the powerful influence a statue such as this had on the continuing Christian
tradition.
In many parish churches in England to this day two panels painted with
Scripture texts hang behind the altar: the ten commandments on the left,
and the Lord’s prayer on the right. This practice goes back to the late
sixteenth century, with even deeper roots in the medieval period. The
setting up of these panels is related to the catechism in the 1552 Book
of Common Prayer. Every person brought before the bishop for confir-
mation was expected to be able to recite the ten commandments and
the Lord’s prayer, as well as the brief explanations of both texts in the
catechism. Perhaps the panels often acted as a crib for those with short
memories!
The impact of this practice on the life of the church is clear. The ten
commandments are singled out as the essence of the law of Moses, while
the Lord’s prayer is assumed to provide the essence of the ‘law of Christ’.
The theological instincts which lie behind the erection of those two panels
of texts are profound, and in my view profoundly correct.
Given the strength of the tradition I have just sketched, it is no surprise
to discover that Gal. 6.2 has often been taken as a reference by Paul to
the importance he attached to the teaching of Jesus. There have been
several influential modern supporters of this interpretation. In his ICC
commentary on Galatians, E. D. Burton wrote: ‘this is one of the few
passages in which the apostle refers to teaching of Jesus transmitted to him
120 Jesus and Gospel
through the Twelve or their companions’.21 Burton, however, failed to list
the Jesus traditions which might have been in the apostle’s mind.
C. H. Dodd was more specific, arguing in 1951 and 1953 that Gal. 6.1-5
was an adaptation of Jesus’ teaching at Matt. 23.4 and 18.15-16.22 At about
the same time Dodd’s pupil W. D. Davies was equally adamant. ‘When he
[Paul] used the phrase n»mov toÓ CristoÓ (the law of Christ) he meant
that the actual words of Jesus were for him a New Torah.’23 More recently,
R. N. Longenecker has defended a more sophisticated version of the same
interpretation. He takes ‘the law of Christ’ to refer to those ‘prescriptive
principles stemming from the heart of the gospel (usually embodied in the
example and teachings of Jesus), which are meant to be applied to specific
situations by the direction and enablement of the Holy Spirit, being always
motivated and conditioned by love’.24
The difficulty with this general line of interpretation is that Paul alludes
very rarely to sayings of Jesus, and refers explicitly to them even less often.25
There is limited evidence in I Corinthians and in Romans 12-14, but even
less evidence in Galatians.26 When the phraseology of Gal. 6.1-5 is compared
closely with the synoptic tradition, there are only two words in common,
b†rh (burdens) and jort©on (load), and even they are not used in similar
contexts.
In my view, interpreters of Gal. 6.2 who claim that the words of Jesus are
‘the law of Christ’ are reading Paul through Matthean eyes. Their reading
may not do justice to Paul, but it does resonate with a long and strong

21 E. D. Burton, The Epistle to the Galatians, ICC (Edinburgh: T. & T. Clark, 1921).
22 John Barclay, Obeying the Truth (Edinburgh: T. & T. Clark, 1988), p. 129 n. 70, astutely notes that
in 1935 C. H. Dodd had denied that Gal. 6.2 could mean the ‘Torah of Jesus’, but had changed his
position by 1951 in his Gospel and Law (Cambridge: Cambridge University Press), pp. 64–83. See also
C. H. Dodd’s important article ‘ ï Ennomov CristoÓ’, reprinted in his More New Testament Studies
(Manchester: Manchester University Press, 1968), pp. 134–8.
23 In his Paul and Rabbinic Judaism (London: SPCK, 1948), p. 144, Davies went somewhat further
than Dodd in claiming that there was rabbinic evidence (albeit somewhat limited) to suggest that
in the new age there would be a new ‘law of the Messiah’. See especially W. D. Davies, The Setting
of the Sermon on the Mount (Cambridge: Cambridge University Press, 1963), pp. 109–90. H. Schlier
also insisted that the ‘law of Christ’ is ‘die Tora des Messias Jesus’, Die Brief an die Galater, KEK,
12th edn (Göttingen: Vandenhoeck & Ruprecht, 1961), p. 272. As there are major problems over the
dating and interpretation of the handful of rabbinic passages Davies and Schlier cite, their case has
won little scholarly support. See especially R. J. Banks, Jesus and the Law in the Synoptic Tradition
(Cambridge: Cambridge University Press, 1975), pp. 65–81.
24 See especially R. N. Longenecker, Galatians, Word Biblical Commentary (Dallas: Word, 1990),
pp. 275–6.
25 For a cautious assessment, see D. C. Allison, ‘The Pauline Epistles and the Synoptic Gospels: The
Pattern of the Parallels’, NTS 28 (1982) 1–32.
26 For Romans see especially M. B. Thompson, Clothed with Christ: The Example and Teaching of Jesus
in Romans 12.1-15.3 (Sheffield: Sheffield Academic Press, 1991).
 The law of Christ and the Gospel 121
tradition which should not be rejected without further ado. We turn now
to a further interpretation of ‘the law of Christ’ with an equally impressive
pedigree, even though it too does not do full justice to Paul’s intention in
Gal. 6.2.

5.5 lex christi, lex amoris

In his first commentary on Galatians (1519), Luther wrote several powerful
and perceptive paragraphs on Gal. 6.2, which he describes as ‘a very beau-
tiful and thoroughly golden maxim: Love is the law of Christ’. In his 1535
commentary on Gal. 6.2, Luther goes even further. ‘The Law of Christ is
the law of love’, he writes three times over. ‘After redeeming and regenerat-
ing us and constituting us as His church, Christ did not give us any new
law except the law of mutual love (John 13.34-5).’ ‘To love does not mean,
as the sophists imagine, to wish someone else well, but to bear someone
else’s burdens, that is, to bear what is burdensome to you and what you
would rather not bear.’
Two centuries after Luther, J. A. Bengel echoed Luther’s interpretation
in a typically pithy comment on Gal. 6.2 in his influential Gnomon Novi
Testamenti (1742): ‘Lex Christi, lex amoris’ (‘the law of Christ is the law of
love’).
The same interpretation is set out much more fully in V. P. Furnish’s
standard textbook The Love Command in the New Testament 27 and sup-
ported by many other writers. It is probably the consensus interpretation
of Gal. 6.2 at present.
Who would want to deny the importance of love in Paul’s ethical
thinking? At Gal. 5.14 love is said to sum up the whole law. Love heads
the list of the fruits of the Spirit at Gal. 5.22, a list which ends with the
ironic comment, ‘there is no law against such things’. Nonetheless, I am
not convinced that this interpretation does full justice to Paul’s intention
in Gal. 6.2, for if the ‘law of Christ’ is taken to mean no more than ‘the law
of love’, Paul’s careful counterpoise with ‘the law of Moses’ is missed, as is
his emphasis on Christ’s own example of self-giving love.

5.6 theological reflect ions

In Gal. 6.2 and I Cor. 9.21 Paul’s Christological emphasis is clear, as is the
love commandment, but there is little or no trace of the teachings of Jesus
as constituting the law of Christ. If we are reflecting on the current value of
27 V. P. Furnish, The Love Command in the New Testament (London: SCM, 1968), pp. 59–65.
122 Jesus and Gospel
‘the law of Christ’ for the Christian Gospel today, why should we privilege
Paul’s usage?
Quite independently of Paul, we find James, Matthew, and the writer of
the Johannine letters using the concept, even if they do not use the actual
phrase. As with Paul, there is a continuing commitment to the law of
Moses as interpreted by Christ. But there is also the addition of a much
clearer emphasis on the teaching of Jesus and a less explicit Christological
note.
Early in the second century, notable theological steps were taken. In the
writings of Hermas, the Kerygma Petri, and Justin Martyr, there is a strong
emphasis on Christ himself as the new law. Arguably one of Paul’s notions
is taken much further, even though it is difficult to trace direct use of Paul’s
letters. In Hermas and Justin, Christ is not only the new law; he is himself
the new lawgiver.
The later Christian tradition develops one or other of the strands of
thought found in the NT and in early Christian writings, but without
adding significant or new understandings. As far as I am aware, the phrase
‘the law of Christ’ is suffering benign neglect in contemporary theological
writing. However, I hope that I have shown that, if we consider Christian
writings up to the middle of the second century and do not confine ourselves
to Paul or to canonical writings, we are not going to be bereft of matters
for continuing theological reflection. Indeed, we have an agenda.
(1) Christology and ethics must not be separated.
(2) The ‘law of Christ’ or whatever alternative or synonymous phrase we
choose must not be allowed to imply abandonment of the law of Moses.
In particular, if we use the phrase ‘new law’, we must do so with care,
for ‘new’ can easily imply that we are discarding the ‘old’. The terms
‘new covenant’ or ‘new testament’ are analogies. They may be used,
but there are lurking dangers. In saying this, I am well aware that there
are unresolved issues concerning Christian theological interpretation
of the Old Testament or the Hebrew Scriptures, if you prefer. But I am
heartened by the number of Biblical scholars who are now wrestling
with these very issues.
(3) The radical, disturbing teaching of Jesus recorded by the evangelists
is of continuing importance, with the love commandment as its focal
point.
(4) In Galatians (and elsewhere) Paul juxtaposes the wisdom of Scripture
(the love commandment of Lev. 19.18 at Gal. 5.14) and baptizes into
Christ the conventional ethical wisdom of the day (bear one another’s
burdens at Gal. 6.2). Why shouldn’t we do likewise?
 The law of Christ and the Gospel 123
(5) Our explorations have warned us yet again not to allow particular
strands of the later Christian tradition to determine our Biblical exe-
gesis. On the other hand, we have seen that later interpretation and
development of a Biblical tradition may stimulate theological reflection
in unexpected directions.
All this seems to me to be a useful starting agenda for a theological
ethics in which Scripture is taken seriously. I am not advocating a return
to the old model whereby the Biblical scholar hands over to the systematic
theologian his or her results, to be used according to whim. Rather, I am
urging that there should be a continuing dialogue between close attention
to Biblical text, the effects of the text (for good or ill) through the centuries,
and contemporary theological reflection.
When we have engaged in this dialogue, we may find that the phrase ‘the
law of Christ’ is too prone to misunderstanding or one-sided interpretation
to be of further use. But the themes associated with this phrase through the
centuries will continue to enrich our theological reflection on the Gospel.
part ii

Jesus
 chap t e r 6

Jesus of Nazareth: a magician and a false prophet

who deceived God’s people?

The relationship of Jesus to first-century Judaism continues to be discussed

vigorously.1 This continuing debate was sparked off initially by the publi-
cation of Hermann Samuel Reimarus’s Wolfenbüttel Fragments between
1774 and 1778.2 In deliberately provocative comments, Reimarus insisted
that Jesus did not intend to abolish the Jewish religion and to introduce
a new one in its place. The intention of Jesus, Reimarus claimed, was
reversed completely after his death by both the actions and the teaching of
the apostles. With their abandonment of the law, ‘Judaism was laid in its
grave.’3
In an equally influential publication two generations later, David
Friedrich Strauss noted that a radical account of the origins of Christianity
along these lines had been propounded by ‘the enemies of Christianity in
its ecclesiastical form’, and that it had been done ‘most concisely of all in
the Wolfenbüttel Fragments’, i.e. by Reimarus.4 Although Strauss was sym-
pathetic to many of Reimarus’s claims, he knew that any presentation of
Jesus as a faithful Jew was built on a one-sided reading of the evidence.
Strauss emphasized that there was clear strong evidence within the gospels
to support the opposite viewpoint: Jesus was at odds with the religious
leaders of his day.5

1 Earlier versions of this chapter were given as the Thatcher Lecture at the Uniting Theological College
in Sydney on 18 July 1991, and as a lecture or seminar paper at the Universities of Otago (New
Zealand), Leiden (the Netherlands), Lancaster, Leeds, Belfast, and Aberdeen. This chapter is a revised
version of my contribution to the I. H. Marshall Festschrift, Joel B. Green and Max Turner, eds.,
Jesus of Nazareth: Lord and Christ (Grand Rapids: Eerdmans, 1994), pp. 164–80.
2 Reimarus died in 1768. The Fragments were published anonymously by G. E. Lessing. The identity
of the author did not become generally known until 1814.
3 Reimarus, Fragments, i.19, ed. C. H. Talbert (Philadelphia: Fortress, 1970), p. 101. In a note at this
point Talbert insists that ‘this entire argument is an oversimplification by Reimarus’.
4 The Life of Jesus Critically Examined, §67. I have cited the edition by P. C. Hodgson (London: SCM,
1973), p. 298; this is George Eliot’s translation of the fourth German edition.
5 Ibid., pp. 297–300 and 599–602.

127
128 Jesus
Reimarus and Strauss both still have plenty of supporters, and many
mediating positions are defended. After one hundred and fifty years of
discussion, the relationship of Jesus to Judaism remains a contentious issue,
as a cluster of influential book titles confirms: Jesus the Jew; Jesus and Judaism;
Jesus and the Transformation of Judaism; Jesus within Judaism.6 Although
our knowledge of first-century Judaism has increased enormously in recent
decades, the variety of views still on offer is bewildering in its rich profusion.
An impasse has been reached. This is partly because at key points there
are crucial gaps in our knowledge. For example, we cannot be certain just
what were the conventions of various religious groups on the fine points of
sabbath observance in, say, Capernaum in ad 30.
And in addition, the relevant evidence which has survived is difficult to
interpret. Almost all of it has come down to us from a partisan point of
view. Whether we are examining the evidence of Josephus, or of the rabbis,
or of the New Testament evangelists, this fact has to be taken very seriously
indeed.
However, I do not think that these difficulties should deter critical in-
quiry. I am convinced that discussion of the relationship of Jesus to con-
temporary Judaism can be advanced as effectively by opening up uncon-
ventional lines of inquiry as it can be by rehearsing familiar arguments. In
this chapter I shall plot a path which has been neglected in most recent
discussion.
I shall examine the most widely attested ancient criticism of Jesus of
Nazareth: he was a magician and a false prophet who deceived God’s peo-
ple.7 The barbed criticisms from the opponents of any striking individual
are often as revealing as the fulsome praise of close associates, so we do
well to take seriously the small number of negative comments about Jesus
which have come down to us from antiquity.8
I shall start with criticisms of Jesus which are found in Jewish, Christian,
and pagan circles in the middle of the second century. I shall then discuss

6 G. Vermes, Jesus the Jew (London: Collins, 1973); E. P. Sanders, Jesus and Judaism (London: SCM,
1985); J. Riches, Jesus and the Transformation of Judaism (London: Darton, Longman & Todd, 1980);
J. H. Charlesworth, Jesus within Judaism (London: SPCK, 1989). In The Historical Jesus: The Life of
a Mediterranean Jewish Peasant (Edinburgh: T. & T. Clark, 1991), J. D. Crossan paints on an even
larger canvas and portrays Jesus as a Mediterranean peasant, a Jewish Cynic. Quotations from the
Cynic letters and Epictetus are given greater prominence than (for example) the Qumran writings.
In effect, Crossan downplays the Jewishness of Jesus. See my review in Theology 95 (1992) 452–3.
7 For a fine discussion of the allegation that Jesus was an ‘illegitimate son’, see S. McKnight, ‘Calling
Jesus Mamzer’, Journal for the Study of the Historical Jesus 1 (2003) 73–103. See also B. D. Chilton,
‘Jésus, le mamzer (Mt 1.18)’, NTS 46 (2000) 222–7.
8 It is worth noting that no ancient opponent of early Christianity ever denied that Jesus existed. This is
the Achilles’ heel of attempts by a few modern scholars such as G. A. Wells to deny that Jesus existed.
 Jesus of Nazareth: a magician and a false prophet? 129
earlier forms of this stock polemic in the writings of Josephus, and then the
four evangelists, before turning, finally, to traditions which go back to the
lifetime of Jesus. It is often helpful in studies of earliest Christianity to work
back from later, clearer evidence to more problematic earlier evidence. Of
course there is a risk of anachronism. But sometimes risks have to be taken:
later traditions may well suggest new questions and new ways of looking
at much-disputed issues.
Accusations of magic and of false prophecy /deception are very closely
related to one another in ancient polemic. In many polemical traditions
in which only one of the two terms occurs, the other is usually implied.
There is a third, closely related concept which is prominent in polemic
in antiquity: the apparent success of the magician and the false prophet /
deceiver is regularly ascribed by opponents to some form of demonic posses-
sion. Readers familiar with elementary mathematics will quickly appreciate
that I have in mind three overlapping circles in a Venn diagram.
I shall try to show that the accusations that Jesus was a magician and a
false prophet who deceived God’s people were well known in the middle
of the second century in both Christian and Jewish circles; they became
part of the stock rabbinic polemic against Jesus. These jibes were probably
known to Josephus about ad 90; they were known to, and countered by, the
evangelists; they were almost certainly used by some of Jesus’ adversaries in
his own lifetime. If this final point can be established, there are important
implications for our appreciation of the relationship of Jesus to the Judaism
of his day.

6 .1 justin mart yr, rabbinic tradit ions, josephus

In the middle of his extended debate with his Jewish adversary Trypho,
Justin claims that the healing miracles of Jesus were the fulfilment of the
messianic prophecies of Isa. 35.1-7. The miracles of Jesus were intended to
elicit recognition of him as Messiah, but many who saw them drew the
opposite conclusion: ‘they said it was a display of magic art, for they even
dared to say that he was a magician and a deceiver of the people (m†gov
kaª laopl†nov)’, Dialogue 69.7. From the context in the Dialogue, there
is no doubt that the term ‘deceiver’ is being used against the background
of Deut. 13.5 in the special sense of a false prophet who leads God’s people
astray.9

9 The notion of ‘deception’ is widespread in ancient polemic of all kinds, where it is not necessarily
linked to false prophecy. False prophecy, however, always involves deception.
130 Jesus
The title of this chapter is taken from this double polemical accusation
in Dialogue 69.7. Justin clearly believes that it was prevalent in the lifetime
of Jesus. Was he correct in this judgement? Or do his comments simply
reflect Jewish–Christian controversies in the middle decades of the second
century? It is not easy to answer these questions – and a great deal is at
stake.
To label someone a ‘magician’ and/or a ‘deceiver’ (‘false prophet’) in
antiquity was an attempt to marginalize a person who was perceived to
be a threat to the dominant social order. If we can show that these terms,
with their roots in Deuteronomy 13 and 18, were used by contemporary
opponents of Jesus, an important corollary will follow: the teachings and
actions of Jesus must have been considered by some of his contemporaries
to be deeply offensive.
The extent to which Justin’s Dialogue reflects genuine discussion and
controversy between Christians and Jews has been keenly debated. The
Dialogue is highly stylized, and it is far from being a dispassionate verbatim
account of a debate between Justin and Trypho. There is at least some
truth in Harnack’s view that Justin’s dialogue with Trypho was in fact the
monologue of a victor.10 However, there is little doubt that the double
accusation against Jesus which is recorded in Dialogue 69.7, ‘magician and
deceiver of God’s people’, was being used in Jewish anti-Christian polemic
in the middle of the second century, for it is also found in two related
rabbinic traditions.11
In b. Sanh. 43a an anonymous tradition is introduced with the formula,
‘It is said’, an indication that it is a baraitha, an old tradition:
On the eve of Passover Yeshu was hanged. For forty days before the execution took
place, a herald went forth and cried, He is going forth to be stoned because he has
practised sorcery and enticed and led Israel astray. Anyone who can say anything in
his favour, let him come forward and plead on his behalf. But since nothing was
brought forward in his favour, he was hanged on the eve of Passover! Ulla retorted:
Do you suppose that he was one for whom a defence could be made? Was he not
a deceiver, concerning whom scripture says [Deut 13.8], Neither shalt thou spare
neither shalt thou conceal him? With Yeshu however it was different, for he was
connected with the government.
10 A. Harnack, Judentum und Judenchristentum in Justins Dialog mit Tryphon (Leipzig, 1930).
11 For a fuller discussion than is possible here, see D. R. Catchpole, The Trial of Jesus (Leiden:
Brill, 1971), pp. 1–71; J. L. Martyn, History and Theology in the Fourth Gospel, 2nd edn (Nashville:
Abingdon, 1979), pp. 73–81; W. Horbury, ‘The Benediction of the Minim and Early Jewish–Christian
Controversy’, JTS 33 (1982) 19–61; J. Maier, Jesus von Nazareth in der talmudischen Überlieferung
(Darmstadt: Wissenschaftliche Buchgesellschaft, 1978); G. Twelftree, ‘Jesus in Jewish Traditions’, in
D. Wenham, ed., Gospel Perspectives: The Jesus Traditions Outside the Gospels, Vol. v (Sheffield: JSOT
Press, 1985), pp. 289–342.
 Jesus of Nazareth: a magician and a false prophet? 131
The pronouncement of the herald, ‘forty days before the execution took
place’, confirms that this is no casual polemical comment, but a formal
legal accusation against Jesus.12 The two verbs ‘entice’ and ‘lead astray’ have
the same direct object, Israel, and are closely related in meaning, as are
the corresponding nouns in the related passage in Mishnah Sanhedrin
7.10-11. Although the comment attributed to Ulla (a late third-century
rabbi) may be a later elaboration of this polemical tradition, the reference
to Deuteronomy 13 confirms that ‘entice’ and ‘lead astray’ amount to a
charge of false prophecy, for in that locus classicus these three terms are very
closely related.
In b. Sanh. 107b the same double accusation is found in the same order:
One day he [R. Joshua] was reciting the Shema when Jesus came before him. He
intended to receive him and made a sign to him. He (Jesus) thinking it was to
repel him, went, put up a brick and worshipped it.13
‘Repent’, said he [R. Joshua] to him. He replied, ‘I have thus learned from thee:
He who sins and causes others to sin is not afforded the means of repentance.’ And
a Master has said, ‘Jesus the Nazarene practised magic and led Israel astray.
These two rabbinic traditions are very difficult to interpret in detail, and
even more difficult to date with any confidence. Were it not for the close cor-
respondence between these traditions and Justin’s Dialogue 69.7, it would
be tempting to dismiss them as third-century (or even later) Jewish anti-
Christian polemic. However, the semi-technical terminology used in Justin
Martyr’s Greek is almost as close as one could reasonably expect to the
Hebrew of the rabbinic traditions.14 Equally important, the order in which
the accusations are cited is the same. In Deuteronomy the discussion of
enticement to apostasy and false prophecy in chapter 13 which is referred
to in b. Sanh. 43a precedes detailed reference to various forms of magic and
sorcery in chapter 18.15 The order of the discussion in Deuteronomy is fol-
lowed in Mishnah Sanhedrin 10-11 (where there is no reference to Jesus), but
it is reversed in Jewish polemic against Jesus both in Justin’s Dialogue and

12 Maier, Jesus von Nazareth in der talmudischen Überlieferung, has claimed that b. Sanh. 43a did not
originally refer to Jesus: that identification was made only in post-Talmudic redaction. W. Horbury,
however, has argued strongly that the sentences ‘on Passover Eve they hanged Jesus’ and ‘Jesus the
Nazarene . . . practised sorcery and deceived and led astray Israel’ may be older than their immediate
context: JTS 33 (1982) 19–61 (57).
13 See especially E. Bammel, ‘Jesus and “Setting up a Brick”’, in his Judaica: Kleine Schriften I (Tübingen:
Mohr, 1986), pp. 204–8.
14 See especially Martyn, History and Theology, p. 79 n. 110.
15 In Deuteronomy 18 rejection of magic in verses 9–14 is followed immediately by the promise of the
‘prophet like Moses’ and discussion of true and false prophecy in verses 15–22. However, the latter
passage does not seem to be in view in either of the two rabbinic traditions quoted above.
132 Jesus
in b. Sanh. 107b and 43a. Since direct dependence is very unlikely, the simi-
lar wording and the correspondence in order strongly suggest independent
use of a stock polemical tradition.
If this double accusation echoed polemical passages in the New Testa-
ment gospels, the close correspondence in terminology and in order might
be coincidental. But this is not the case. Jesus is not called ¾ m†gov (‘magi-
cian’) in the New Testament. Although the pl†nov (‘deceiver’) accusation
occurs explicitly (as we shall see) in redactional passages in the gospels, there
are no grounds for suspecting literary dependence. So the jibes against Jesus
which are found both in Justin’s writings and in rabbinic traditions have
not been taken directly from the canonical gospels. They put us in touch
with an independent negative assessment of Jesus which appears to have
been widespread by the middle of the second century, and probably much
earlier.
There is a further example of this double accusation in the Acts of Thomas,
a third-century writing which almost certainly incorporates earlier tradi-
tions. In a strongly polemical passage in chapter 96, Charisius rounds on
his wife Mygdonia: ‘I have heard that that magician and deceiver (¾ m†gov
–ke±nov kaª pl†nov) teaches that a man should not live with his own wife.’
He said to her again, ‘be not led astray by deceitful and vain words nor
by the works of magic’. The phrase ‘magician and deceiver’ recalls Justin’s
Dialogue 69.7 and the rabbinic allegations against Jesus. The same phrase
is used in chapter 102, and again (though in separate sentences) in chapters
106–7.16
These passages in the Acts of Thomas are very different from the examples
of this allegation which have been referred to above. Here there is hardly a
trace of Christian–Jewish polemic or apologetic, and, with the exception of
chapter 48 (where Jesus is referred to with the single accusation, ‘deceiver’,
pl†nov), they are made against the apostle Thomas, not against Jesus. But
there is no doubt at all that Thomas is an alter ego of Jesus.17 The use of the
phrase ‘magician and deceiver’ (m†gov kaª pl†nov) can hardly be a coinci-
dence. In the Acts of Thomas a ‘stock’ Jewish criticism of Jesus has survived

16 The English translation is taken from E. Hennecke, New Testament Apocrypha, Vol. ii, ed. R. Wilson
(London: Lutterworth, 1965). The Greek text is from Acta Apostolorum Apocrypha, ii/2, ed. M. Bonnet
(Leipzig, 1903; repr. Hildesheim, 1959). The reference in chapter 48 to Jesus as ¾ pl†nov is noted,
but without discussion, by Martyn, History and Theology, p. 79 n. 110, and by A. Strobel, Die Stunde
der Wahrheit (Tübingen: Mohr, 1980), p. 90, but they do not refer to chs. 96 or 106–7. A. F. J. Klijn,
The Acts of Thomas (Leiden: Brill, 1962), p. 271, includes a note on the ‘sorcerer’ accusation which is
made against Thomas in numerous passages, but does not comment on the pl†nov accusation or
on the combination of m†gov and pl†nov noted above.
17 See, for example, chapters 2, 11, and 45.
 Jesus of Nazareth: a magician and a false prophet? 133
in a very different setting. This confirms that it was a widespread and
well-known polemical accusation.

A further example of this double accusation is found in the Testimonium

Flavianum, the paragraph about Jesus in Josephus, Antiquities xviii.3. In its
present form it cannot have been written by the Jewish historian c. ad 93–4,
for several phrases are undeniably Christian assessments of the significance
of Jesus. The key question, debated since the sixteenth century, is whether
the whole five-sentence paragraph is a later Christian interpolation, or
whether a Christian scribe has added a few phrases (and perhaps altered
a few words) in a ‘neutral’ or mildly hostile account of Jesus which was
originally written by Josephus himself.18
Over the last decade or so, several influential voices have supported the
latter view.19 I believe that this is much the more likely solution. For our
present purposes the second sentence is the most important. In the Loeb
edition of the writings of Josephus, L. H. Feldman translates it as follows:
‘For he [Jesus] was one who wrought surprising feats and was a teacher of
such people as accept the truth gladly. He won over many Jews and many
of the Greeks.’ Translations along these lines are found in several books on
Jesus and the gospels; they offer a neutral or even mildly positive assessment
of Jesus. I believe that they are unduly influenced by the view that the
whole pericope is a later Christian addition. Without any emendation of
the text, the key Greek words can be translated more plausibly to give an
ambivalent or even a mildly hostile assessment of Jesus, i.e. one which can
be attributed to Josephus with much greater confidence.
Let us start with the final verb, –phg†geto, translated by Feldman as
‘he won over’, and by J. P. Meier as ‘he gained a following among’. R. Eisler
emended the verb to ˆphg†geto, and translated it as ‘seduce’.20 If this
proposal is adopted, we are close to Justin’s laopl†nov (‘deceiver of the
people’) and the related rabbinic terms discussed above.

18 See now James Carleton Paget’s important detailed discussion, ‘Some Observations on Josephus and
Christianity’, JTS 52 (2001) 539–624.
19 See especially E. Bammel, ‘Zum Testimonium Flavianum (Jos Ant 18, 63-64)’, in O. Betz, K. Haacker,
and M. Hengel, eds., Josephus-Studien, FS O. Michel (Göttingen: Vandenhoeck & Ruprecht, 1974),
pp. 9–22, reprinted in his Judaica, pp. 177–89. For a perceptive assessment of the literature, see J. P.
Meier, ‘Jesus in Josephus: A Modest Proposal’, CBQ 52 (1990) 76-103. Meier argues that, with the
extraction of the three most obviously Christian statements, the Testimonium yields the original or
‘core’ text Josephus wrote, with no need to rewrite any words or phrases in the core. I believe that
this solution is correct, but I do not accept all Meier’s translations of what is admittedly difficult and
ambiguous Greek. Meier does not discuss the relationship of the original text of the paragraph to
the other assessments of Jesus with which I am concerned in this chapter.
20 R. Eisler, The Messiah Jesus and John the Baptist (London, 1931), pp. 61–2.
134 Jesus
But there is no need to emend the verb. The Bauer–Danker lexicon
gives ‘to cause a state or condition to be or occur’, ‘to bring on’ as the
meaning of –p†gw, and notes that it usually has the sense ‘bring some-
thing bad upon someone’.21 Hence –phg†geto in the Testimonium can
be understood as ‘brought trouble to’, or even ‘seduced, led astray’.22 In
other words, the verb –p†gomai is only a little less close to plan†w than is
ˆp†gomai.
The preceding two phrases in the Testimonium are equally important for
our present concerns. The translation of the first is not problematic: the
phrase refers to Jesus as ‘one who did surprising (or unexpected) deeds’.
Depending on one’s perspective, this could refer negatively to a magician,
or positively to a miracle worker.
Meier translates the next phrase as ‘a teacher of people who receive the
truth with pleasure’, but in a note accepts (correctly) that it could imply
simple-minded enthusiasm, even self-delusion.23 A decision depends on
whether Josephus paints his portrait of Jesus from a mildly hostile or from
a ‘neutral’ perspective. I believe that the former is much more likely, though
a full defence is not possible here.24
In short, in the Testimonium Jesus is said to have been a miracle worker /
magician who impressed rather gullible people, and led many Jews (and
many Greeks) astray. Although the terminology in the terse assessment of
Jesus in the Testimonium differs from that used in the anti-Christian Jewish
polemic quoted by Justin, and in the rabbinic traditions discussed above,
there is notable agreement.
In all these summary assessments of Jesus, his actions and his words are
linked closely; they are referred to in the same order, and in broadly similar
ways, as those of a ‘magician’ and a ‘false prophet’ who led God’s people
astray. I do not think that these similarities can be mere coincidence. From
these passages, which are all independent of one another, we may conclude
that there was a stable form of anti-Jesus polemic which may date from the

21 Josephus, Life 18 is a good example of the verb in this sense. ‘Win over’ is attested in Thucydides
and Polybius (see Liddell, Scott, and Jones, Lexicon) and Chrysostom (see Lampe, Patristic Greek
Lexicon), but the verb is rarely used with this positive sense.
22 E. Bammel notes that significatio seditionis is possible for –p†gomai; ‘Zum Testimonium Flavianum’,
Judaica, pp. 179–81. J. P. Meier acknowledges that this is ‘a possible though not necessary meaning
of the verb’, but does not give supporting references or reasons for rejecting this translation: ‘Jesus in
Josephus’, p. 88 n. 33. M. Smith, Jesus the Magician (London: Victor Gollancz, 1978), p. 178 translates
‘lead astray’ and claims that this sense is implied by the Greek text.
23 ‘Jesus in Josephus’, p. 84 and n. 19.
24 See especially Bammel, ‘Zum Testimonium Flavianum’, Judaica, pp. 177–89. Bammel’s article is per-
ceptive and generally persuasive, though I am not convinced by his proposed conjectural emendation
of ˆgapžsantev to ˆpatžsantev.
 Jesus of Nazareth: a magician and a false prophet? 135
time of Josephus towards the end of the first century. In the final section
of this chapter we shall consider its earlier roots.

6 . 2 m ag ic, false prophecy, demonic possession

So far we have considered examples of the double polemical accusation that
Jesus was a magician and a false prophet / deceiver. Both accusations are
also found singly. Since, as we shall see in a moment, these charges are
closely related to one another, use of one or other polemical term often
carries with it an implication of the other.
The single accusation that Jesus was a sorcerer is found in several passages.
In I. Apol. 30 Justin refers to the claim by his opponents that the miracles
of Jesus were done by the use of magical arts. This accusation is also made
repeatedly by Celsus, writing at about the same time as Justin, usually when
he is citing the polemic of a Jew.25
The single accusation that Jesus was a false prophet / deceiver is found
even more widely. In chapter 108 of his Dialogue, Justin claims that Jewish
leaders ‘appointed chosen men and sent them into all the civilized world,
proclaiming that “a certain godless and lawless sect has been raised by
one Jesus of Galilee, a deceiver (pl†nov)”’. A Christian redaction of The
Testament of Levi, which may well date from the same time as Justin and
Celsus, alleges that those who plotted to kill Jesus categorized him as a
deceiver (pl†nov) (T. Levi 16.3).26 As we shall see in the final section of
this chapter, this line of polemic is also found in the gospels.
The following passages confirm the close relationship in antiquity be-
tween ‘magician’, ‘impostor’ (m†gov, g»hv) and ‘false prophet / deceiver’
(yeudoprojžthv / pl†nov).
(1) At Acts 13.6-12 Luke records that Barnabas and Paul met a certain
magician, a Jewish false prophet named Bar-Jesus, also known as Elymas
(tin‡ yeudoprojžthn m†gon, verse 6). Luke carefully contrasts Paul, who
has been filled with the Holy Spirit, with Elymas, who is a son of the devil
and full of deceit and villainy (verses 9–10). In this passage ‘magician’ and
‘false prophet’ are almost synonymous; not quite synonymous, however,
for otherwise one or other term would be redundant.

25 See Contra Celsum i.6, 28, 68, 71; ii.32, 48-9. On Celsus’ Jew, see Bammel, ‘Der Jude des Celsus’, in
his Judaica, pp. 265–83.
26 For a fuller discussion of these passages, see G. N. Stanton, ‘Aspects of Early Christian–Jewish Polemic
and Apologetic’ NTS 31 (1985) 377–82, included in A Gospel for a New People: Studies in Matthew
(Edinburgh: T. & T. Clark, 1992), pp. 232–55. See also The Apocryphon of John 1.1; Origen, Contra
Celsum ii.1; Chrysostom, Sermons against Judaizers, v.5.8-9.
136 Jesus
(2) In Philo, Spec. Leg. i.315, ‘prophet’ and ‘impostor’ (projžthv and
g»hv) are contrasted, probably with the discussion of false prophecy in
Deuteronomy 13 in mind:
If anyone cloaking himself under the name and guise of a prophet and claiming
himself to be possessed by inspiration lead us on to the worship of the gods
recognised in the different cities, we ought not to listen to him and be deceived
by the name of a prophet. For such a one is no prophet, but an impostor (g»hv),
since his oracles and pronouncements are falsehoods invented by himself.

In this passage g»hv is used in the general sense of ‘impostor’, ‘charlatan’;

‘sorcery’ is probably, but not necessarily, in view. However, ‘false prophecy’
and ‘sorcery’ are associated explicitly in Vit. Mos. i.277, where Philo con-
trasts the technique of the magician (m†gov) with the ‘prophetic spirit’
(pneÓma projhpk»n): ‘there fell upon him [Balaam] the truly prophetic
spirit which banished utterly from his soul his art of wizardry’; Balaam then
delivers a prophetic oracle.27
(3) In Josephus, Ant. 20.169-72; Bell. 2.261-3 (cf. Acts 21.38), we read of an
unnamed Egyptian who designated himself as a prophet (projžthv) and
promised a legitimating miracle: at his command, Jerusalem’s walls would
fall down. Josephus, however, scornfully labels him as a ‘false prophet’ and
‘sorcerer’ (yeudoprojžthv and g»hv).28
In Ant. 20.97 Josephus refers to a certain sorcerer (g»hv) called Theudas,
who persuaded a large number of people to follow him to the river Jordan;
‘he stated that he was a prophet and that at his command the river would
be parted and would provide him with an easy passage’. Josephus clearly
believes that Theudas was a false prophet whose promised miracle was a
typical ploy of a g»hv.
(4) This close relationship between ‘magic’ and ‘false prophecy’ is by
no means confined to Hellenistic Jewish and early Christian writers. A
second century ad papyrus notes that, in the eyes of his opponent, the
prophet of Apollo is a ‘hungry g»hv’.29 In Philostratus, Life 5.12 we read
that Apollonius does not prophesy on the basis of ‘sorcery’ (gohte©a),
but on that of divine revelation. Lucian of Samosata, Peregrinus 13 attacks
opponents as ‘false prophets who are really charlatans and sorcerers’. And

27 See G. Delling, art. m†gov in TDNT iv, p. 358. In most of the passages with which I am concerned,
there is no significant difference between m†gov and g»hv.
28 See also the descriptions of individual rebels in Josephus, Ant. 20.97-8; 18.1-10 (and 20.102); and sum-
mary passages in Ant. 20.167f.; Bell. 2.259. For discussion, see D. Aune, ‘Magic in Early Christianity’,
ANRW ii.23.2, ed. W. Haase (Berlin: de Gruyter, 1980), pp. 1540–1, here p. 1528.
29 For details, see BDAG, sub g»hv.
 Jesus of Nazareth: a magician and a false prophet? 137
in an intriguing magical papyrus which is difficult to date (PGM 5.110), we
read, ‘I am Moses your prophet.’30
These passages (which are by no means exhaustive) confirm the close
relationship between ‘magic’ and ‘false prophecy’. In ancient polemic, refe-
rence to one or other term often implies the other.31 From the evidence set
out above, it is clear that the allegations that Jesus was a magician and that
he was a false prophet who deceived God’s people are closely related. They
are found in tandem (and, strikingly, in the same order), and also singly in a
wide range of writings. The strength of the evidence from the middle of the
second century is most impressive. The Testimonium Flavianum suggests
that the polemic has deeper roots in the first century.
Before we explore the origin of what became stock polemic against Jesus
and early Christianity, it will be important to show that there is a third,
closely related line of polemic in writings from our period. The examples
which follow show that, in ancient polemic, opponents often allege that
both the ‘magician’ and the ‘false prophet’ are able to act as they do as the
result of their close relationship to the devil or to demons.
(1) At the beginning of chapter 69 of Justin’s Dialogue with Trypho, the
chapter which I have taken as my starting point in this chapter, Justin notes
that both the deeds of the magicians in Egypt and the words of the false
prophets in the time of Elijah were the work of the devil. This linking
of magic, false prophecy, and the work of the devil is significant. It is
found in several passages in Justin and in a wide range of ancient polemical
writings.
A further important example of the close relationship of these three
concepts occurs at the opening of the Dialogue. When Justin recounts his
conversion experience to his Jewish dialogue partner Trypho, he attaches
great weight to the witness of the Old Testament prophets. The fulfilment
of their prophecies compels agreement with what they have spoken:
And also on account of the miracles which they did, they were entitled to belief,
for they both glorified the Maker of all things as God and Father, and proclaimed
the Christ sent from Him, as His Son, a thing which the false prophets who are

30 I owe this reference to Anitra Bingham Kolenkow, who quotes the passage in full: ‘Relationships
between Miracle and Prophecy in the Graeco-Roman World and Early Christianity’, ANRW ii.23.2,
ed. W. Haase (Berlin: de Gruyter, 1980), p. 1488.
31 In Jesus the Magician, p. 79, Morton Smith goes further and states that ‘false prophet’ and ‘magician’
were often used almost as synonyms. However, in his Appendix B, ‘Jesus vs. the Prophets’, he
downplays evidence which suggests that Jesus saw himself and was seen by others as a prophet.
P. Samain has also stressed the close relationship between ‘impostor’ and ‘magician’: ‘L’Accusation
de magie contre le Christ dans les évangiles’, ETL 15 (1938) 449–90. I have not been able to consult
this article; see the summary in Aune, ‘Magic in Early Christianity’, pp. 1540–1.
138 Jesus
filled with the seducing and unclean spirit never did nor even do, but dare to work
miracles of a sort to amaze men, and give glory to the spirits of error and demons.
(Dialogue 7.3)32

(2) About fifteen years after Justin wrote his Dialogue (c. ad 160), the
pagan philosopher Celsus mounted the oldest surviving literary attack on
Christianity, known to us only from Origen’s reply, Contra Celsum. In a
number of passages Celsus quotes approvingly the objections of a Jew. At
a key point Celsus’ Jew challenges Jesus: ‘Come let us believe that these
miracles were really done by you’. He then refers to the works of sorcerers
who profess to do wonderful miracles and asks, ‘Since these men do these
wonders, ought we to think them sons of God? Or ought we to say that
they are the practices of wicked men possessed by an evil daemon?’33
There is no doubt or even hesitation on the part of Celsus’ Jew about
miracles. The key question is whether the ‘wonderful miracles’ of Jesus were
the result of demonic possession or of a special relationship to God.
(3) In his Panarion 48.1-13 Epiphanius quotes at length an anti-Montanist
source. This includes a warning that disciples of the prophetess Maximilla
(associated with Montanus) may well be shown to be false prophets inspired
not by the Holy Spirit, but by demonic error, and to have duped their
hearers.
(4) Within the New Testament there are several examples of the close
relationship between false prophecy, magic, and demonic possession which
was widely accepted in antiquity.
(a) In a Q tradition (Matt. 11.7-19 = Luke 7.24-35) John the Baptist is
accepted by Jesus as a prophet, indeed more than a prophet. Yet John is
written off by ‘this generation’ as a false prophet, with the jibe, ‘he has a
demon’.
(b) Luke’s carefully drawn contrast in Acts 13.6-12 between Paul and
Bar-Jesus / Elymas (noted above) is striking. Paul has been filled with the
Holy Spirit, but Elymas is a son of the devil and full of deceit and villainy
(verses 9–10). Whereas the true prophet acts and speaks as a result of being
filled with the Spirit, the magician and false prophet Bar-Jesus / Elymas is
in league with the devil.
32 With some minor modifications I have quoted A. L. Williams’s translation: Justin Martyr: The
Dialogue with Trypho (London: SPCK, 1930). For a further example from Justin, see I Apol. 26
and 56: Menander, a Samaritan and a disciple of Simon Magus, ‘was possessed by the demons. He
deceived many at Antioch by magic arts.’ (ch. 26).
33 Contra Celsum i.68. I have quoted the translation by H. Chadwick, Contra Celsum (Cambridge:
Cambridge University Press, 1953). See also Contra Celsum ii.49, where Celsus’ Jew uses a similar
line of argument: Jesus referred to miracle workers whom he clearly regarded as wicked men and
sorcerers (Matt. 24.23-7 and 7.22-3), so why should one conclude that the miracles of Jesus are those
of a god, while those who employ similar miracles are sorcerers?
 Jesus of Nazareth: a magician and a false prophet? 139
(c) In I John 4:1 false prophets do not confess that Jesus is from God, for
they are inspired by the spirit of the antichrist.
(d) In Rev. 2.20-5 the false prophetess Jezebel’s outrageous actions and
deeds are linked to her teaching ‘the deep things of Satan’.
(e) In Rev. 13.11 we meet the second beast, who is referred to as a ‘false
prophet’ three times (16.13; 19.20; 20.10). John Sweet astutely notes that
its ‘great signs’ and ‘fire from heaven’ (13.13) parody the activity of the
Pentecostal Spirit and the true prophets.34 The false prophet’s signs deceive
(plan†w) those who had received the mark of the beast (13.14 and 19.20).
In Rev. 16.13-14 three foul spirits like frogs issue from the mouth of the false
prophet (and the dragon and the beast); all three are demonic spirits who
perform signs. In other words, the ‘signs’ of the false prophet are the result
of demonic possession.

In the first and second centuries ad, three closely related concepts are found
in polemical traditions which attack Jesus and a wide range of other indi-
viduals. General accusations of deception are often given sharper focus by
reference to false prophecy. In Jewish polemic which echoes Deuteronomy
13, this is nearly always the case: the ‘deceiver’ is a ‘false prophet’ who leads
God’s people astray.
Although g»hv is used in the more general sense of ‘rogue’ or ‘impostor’,
where the individual concerned has laid claim to legitimating miracles (and
this is the norm, rather than the exception), the label g»hv implies sorcery,
as, of course, does m†gov.35
The claim that an opponent was possessed by demons or in some other
way was closely related to the demonic world was easy to make and difficult
to refute. As we have seen, it was often used in conjunction with allegations
of sorcery and false prophecy. These three labels, which are so prominent in
ancient polemic, have a specific social setting. They are used to marginalize
and undermine the influence of individuals whose claims and behaviour are
perceived to pose a threat to the stability of the dominant social order. In
short, the polemic is a form of social control.36 With these considerations
in mind, we turn now to the gospels.

34 John Sweet, Revelation, Pelican Commentaries (London: SCM, 1979), p. 214.

35 In the writings to which I have referred, with the exception of Matt. 2.1-16, m†gov always has a
pejorative sense.
36 For a helpful discussion of the social setting of magic, see Aune, ‘Magic in Early Christianity’,
pp. 1510–16. Part of L. T. Johnson’s article, ‘The New Testament’s Anti-Jewish Slander and the
Conventions of Ancient Polemic’, JBL 108 (1989) 419–41, is relevant at this point. See also Bruce J.
Malina and Jerome H. Neyrey, Calling Jesus Names: The Social Value of Labels in Matthew (Sonoma,
Calif.: Polebridge, 1988), esp. pp. 35–42.
140 Jesus
6.3 the evangelists
Before considering the possibility that already in his own lifetime Jesus
was labelled by his opponents as a magician and a false prophet, we must
examine evidence which stems from the hands of the evangelists. This
procedure is in line with the method of ‘working backwards’ from later,
clearer evidence to earlier, more problematic evidence, which I have used
in earlier sections of this chapter. I take this method to be axiomatic for all
critical inquiry into the teaching and actions of Jesus.37
In the central chapters of John’s Gospel we are able to overhear the
disputes between the Johannine community and the local synagogue in
the evangelist’s own day. In chapter 7 there are three references to division
among the people on account of Jesus (7.12, 25-7, 40). In 7.12 some who are
antagonistic to Jesus claim that he is leading the people astray. At the climax
of the chapter, the officers who were sent by the chief priests and Pharisees to
arrest Jesus returned empty-handed, only to be asked why they had failed to
bring Jesus with them (verse 45). The officers answered, ‘No man ever spoke
like this man!’ (verse 46). Their positive response to Jesus is immediately
undermined by the taunt, ‘Are you led astray, you also?’ (verse 47).
In both 7.12 and 7.47 reference to ‘leading astray’ (plan†w) is to
a formal allegation, with roots in the discussion of false prophecy and
apostasy in Deuteronomy 13. In John 4.19 and 9.17 acknowledgement of
Jesus as a prophet is shown to be an acceptable, if partial, response to
Jesus. So it is no surprise to find reference to Jesus as a false prophet in 7.12
and 7.47 and, by implication, in 7.52. There is little doubt that the jibe is
related to the polemic Justin refers to in Dialogue 69.7 and 108.2, and to
the allegations found in the rabbinic traditions quoted above. In both 7.12
and 7.47 the claim that Jesus leads the people astray is used to ridicule
sympathetic responses to Jesus. This may well have been the context in
which this jibe was used in the evangelist’s day. The evangelist refers to
this and to other allegations against Jesus because he is confident that his
gospel as a whole is an adequate response.
The claim that Jesus is a magician is not referred to explicitly in
the Fourth Gospel. Perhaps this is not surprising, given the absence of
references to exorcism in this gospel and the fact that exorcism is unques-
tionably the best-attested form of magic among Jews before Bar Kokhba.38

37 I have tried to use this method consistently in my book, The Gospels and Jesus.
38 So P. S. Alexander, ‘Incantations and Books of Magic’, in E. Schürer, The History of the Jewish People
in the Age of Jesus Christ iii.1, ed. G. Vermes, F. Millar, and M. Goodman (Edinburgh: T. & T. Clark,
1986), §32.vii, pp. 342–79, here p. 342.
 Jesus of Nazareth: a magician and a false prophet? 141
Although John 8.48, which contains the allegation that Jesus has a demon,
has sometimes been said to be related to the charge that he was a magician,39
this is unlikely. We noted above that allegations of demonic possession were
associated with charges both of sorcery and of false prophecy. Since the
latter is so clearly in view in John 7, and since in the immediate context it is
the (prophetic) words of Jesus which are being attacked by his opponents,
false prophecy rather than sorcery probably lies behind John 8.48.
John 10.19-21 is all of a piece.40 As in John 7.12 and 47, following a
division among the people, a hostile allegation is used to discredit those
who make a sympathetic response to Jesus. Jesus is said to have a demon
and therefore to be mad. ‘So why listen to him as the prophet like Moses
(to him you shall listen, Deut. 18.15)?’ Others insist that the sayings of Jesus
are not those of one who has a demon. And then, unexpectedly for modern
readers, they say, ‘Can a demon open the eyes of the blind?’ Although this
final question might be taken as a rebuttal of an accusation that Jesus is
a magician, once again (as in John 8.48) false prophecy is more likely to
be in view. As we noted above, demon possession is often said to confirm
false prophecy as well as sorcery. And there are some Jewish traditions in
which the prophet like Moses is expected to perform signs.41 This hope is
reflected in John 6.14; 7.31; 9.16-17 and, we may add, 10.21.
In view of the role which accusations of demonic possession (and similar
charges) play in ancient polemic, it is perhaps not surprising that both
parties to the ferocious disputes which lie behind the central chapters of the
Fourth Gospel use this taunt. In John 8.44 ‘the Jews’ are said to be ‘of their
father the devil’, while in 8.48, 52, and 10.20 Jesus is alleged to have a de-
mon. The evangelist is confident that his readers will know where the truth
lies.
In Luke 23 there are three references to Jesus leading the people astray.
In verse 2 the verb is diastr”jw, in verse 5 ˆnase©w, and in verse 14
ˆpostr”jw. These clauses are certainly not taken from Matthew or from
John; in all probability they come from Luke’s own redaction of Marcan
traditions rather than from pre-Lucan material. From Luke’s perspective,
such charges were mischievous if not unexpected; they were all of a piece

39 J. L. Martyn, History and Theology in the Fourth Gospel, 2nd edn (Nashville: Abingdon, 1979), p. 77,
notes in passing that this is a possibility. See also R. Bultmann, The Gospel of John (E. tr. Oxford:
Blackwell, 1971), p. 299 n. 4.
40 The importance of these verses has been regularly overlooked in the standard commentaries and
recent major books on the Fourth Gospel.
41 See Martyn, History and Theology in the Fourth Gospel, pp. 106–12, with references both to primary
sources and to secondary literature.
142 Jesus
with the false allegations brought against Stephen and Paul.42 These three
verses in Luke 23 strongly suggest that Luke was aware of what became the
standard polemical claim that Jesus was a false prophet who led the people
astray. This is made more probable by the comment in the next chapter,
made by two followers of Jesus on the road to Emmaus: they acknowledge
Jesus as a prophet mighty in deed and word (24.19). So in the finale of
his story the evangelist poses the stark alternative: was Jesus the expected
messianic prophet who would redeem Israel (24.19, 21, 26) or was he a
false prophet who was deceiving God’s people?
Matthew’s Gospel reflects the double allegation that Jesus was a magician
and a false prophet who deceived God’s people. The threefold accusation
that the exorcisms of Jesus have been carried out ‘by the prince of demons’
(9.34; 10.25; 12.24, 27) is a way of alleging that Jesus is a magician. As we
noted above, exorcism is the best-attested form of magic among Jews before
Bar Kokhba.43 The first two of Matthew’s three references to this accusation,
9.34 and 10.25, come from the evangelist’s own hand. The third, 12.24, 27,
is taken from Mark 3.22 and from Q (= Luke 11.19); Matthew has redacted
his traditions at this point. So the evangelist Matthew clearly has a special
interest in this allegation: he is anxious to acknowledge it and to refute it.
At the climax of Matthew 12 the tables are turned on the scribes and
the Pharisees: they are part of a generation which is possessed by seven evil
spirits (12.43-5). They are demon possessed, not Jesus. As in the Fourth
Gospel, both sides in this bitter dispute between Christians and Jews trade
the same taunt.44 The threefold accusation that the exorcisms of Jesus
have been carried out by dint of collusion with Beelzebul, the prince of
demons, is carefully balanced by a threefold insistence that Jesus acts ‘by
the Spirit of God’ (–n pneÅmati qeoÓ; 12.18, 28, 31-2).
The final words attributed to the Jewish leaders in Matthew refer to
the second half of the double allegation with which we are concerned. In
27.63-4 Jesus is referred to as ‘that deceiver’ (–ke±nov ¾ pl†nov) and his life
is summed up as ‘deception’ (pl†nh). Once again the Pharisees are singled
out as the arch-opponents. They have been conspicuously absent from
Matthew’s story line since the end of chapter 23, but in 27.62-6 they join
the chief priests (whose presence is demanded by the preceding narratives)
in petitioning Pilate. The whole pericope is thoroughly Matthean, so here

42 See G. N. Stanton, ‘Stephen in Lucan Perspective’, in E. A. Livingstone, ed., Studia Biblica III
(Sheffield: JSOT Press, 1980), pp. 345–60.
43 In Acts 19.11-20 Luke almost equates exorcism and magic. See also Josephus, Ant. viii.45-9; Justin,
Dialogue 85.3; Origen, Contra Celsum i.68.
44 For a fuller discussion of these passages, see Stanton, A Gospel for a New People, pp. 173–9.
 Jesus of Nazareth: a magician and a false prophet? 143
we have further evidence of the evangelist’s own special interest in a hostile
assessment of Jesus.45
This time Matthew does not reply directly to the polemic. He takes
great pains to convince the reader that the resurrection of Jesus from
the tomb in which he was buried was not the ‘final deception’, but he
simply lets the Jewish leaders’ critical comments stand. Presumably he is
convinced that readers of his gospel will readily agree that the claim of the
Jewish leaders that Jesus is a ‘deceiver’ is monstrous; perhaps the closing
verses of the gospel (28.18-20) were intended to prove the point.
And so to our earliest gospel, Mark. Chapter 3.19b–35 is a classic example
of the evangelist’s ‘sandwich technique’ of intercalation. The evangelist’s
own redactional hand is evident at several points.46 The response of
Jesus to the strong attack launched by the scribes from Jerusalem (verses
22–30) is inserted into the response of Jesus to the only marginally less
hostile comments of his family (verses 19b–21 and 31–5). The criticisms are
strikingly similar. The family of Jesus accepts the validity of the crowd’s
conviction that Jesus was mad – and in antiquity madness was generally
ascribed to possession by an evil spirit. The scribes from Jerusalem bring
a double charge: Jesus is possessed by Beelzebul (verse 22a); by the prince
of demons he casts out demons (verse 22b).
As in the other examples of Mark’s distinctive sandwich technique, the
intercalated tradition and the framework into which it is inserted interpret
and reinforce one another. In this case the evangelist is acutely aware
of the jibe that Jesus is a demon-possessed magician, and responds to it
vigorously by means of the sayings of Jesus in verses 23b–29 and in verses
33–5. The fact that what I take to be the same jibe is repeated four times
in this passage is striking (verse 21 family / crowd; verse 22 bis the scribes;
verse 30 the evangelist’s redactional repetition of the scribes’ attack).
The allegation that Jesus was a magician and that he was a false prophet
was known at the time the evangelists wrote. Matthew knew the double
form of the accusation. John (certainly) and Luke (very probably) were
aware that Jewish opponents of Christian claims alleged that Jesus was a
false prophet who led God’s people astray. Mark is stung by the strength
of the claim that Jesus’ actions, and in particular his exorcisms, were the
result of demon-possession.

45 For details of Matthean vocabulary and style in this pericope, see R. H. Gundry, Matthew: A
Commentary on his Literary and Theological Art (Grand Rapids: Eerdmans, 1982), pp. 540–1.
46 See the fine discussion in S. C. Barton, Discipleship and Family Ties in Mark and Matthew (Cambridge:
Cambridge University Press, 1994), pp. 68–86.
144 Jesus
The passages from the gospels discussed in this section all bear the stamp
of the individual evangelists. They confirm that the evangelists were all
aware of polemical allegations against Jesus and sought to counter them.
The four evangelists all wrote in the 70s and 80s, at a time of mutual
incomprehension, keen rivalry, and sour disputes between Christians and
Jews. So it is no surprise to find in all four gospels charges and counter-
charges concerning the significance of the actions and teaching of Jesus.

6.4 jesus
We have now traced anti-Jesus polemic back from the latter half of the sec-
ond century ad to the time when the four gospels were written, in the 70s
and 80s of the first century. Does this polemic have earlier roots in the life-
time of Jesus? I shall argue that there are sound reasons for concluding that
even in his own lifetime Jesus was labelled ‘a magician’ by his opponents,
and very probably ‘a false prophet’ who led God’s people astray.
First of all we shall return to the double criticism of the scribes from
Jerusalem in Mark 3.22: ‘He has Beelzebul, and by the ruler of demons he
casts out demons.’ The opponents of Jesus represent the dominant social
order which he is threatening.47 Although the reference to Jerusalem may be
a redactional note from Mark himself, the scribes were probably identified
as opponents of Jesus in the pre-Marcan tradition.
The same jibe is recorded independently in a Q tradition: ‘It is only by
Beelzebul, the ruler of the demons, that this fellow casts out the demons.’48
In Matthew’s version (12.24) of the Q tradition it is said to stem from
the Pharisees, and, in Luke’s (11.15), from some of the crowd. The varied
attribution of the criticism is striking. This is a stock taunt thought to have
been thrown at Jesus by more than one group of opponents.
The content of the jibe is stable: in both Mark and in Q the exorcisms of
Jesus are said to be carried out as a result of Jesus’ association with Beelzebul,
the prince of demons. We noted above that exorcism was the best-attested
form of magic among Jews in the first century and that magicians were
regularly said to be demon-possessed. Hence there is little doubt that both
47 A. F. Segal’s comments are instructive: Rebecca’s Children: Judaism and Christianity in the Roman World
(Cambridge: Harvard University Press, 1986), pp. 144–5: ‘The logic from the scribes’ perspective is
that if Jesus were from God, he could not oppose the ideas of the legitimate authorities of Judea.
Since he does oppose them, his power must have other sources.’
48 In most recent reconstructions of Q, Matt. 12.24 = Luke 11.15 is accepted as a Q tradition. See
especially David Catchpole, The Quest for Q (Edinburgh: T. & T. Clark, 1993), pp. 48–50; also J.
Kloppenborg, Q Parallels: Synopsis, Critical Notes & Concordance (Sonoma, Calif.: Polebridge, 1988),
pp. 90–2.
 Jesus of Nazareth: a magician and a false prophet? 145
the Marcan and the Q traditions are tantamount to a charge that Jesus was
a magician.
In his healing miracles and exorcisms Jesus undoubtedly used techniques
which would have been perceived by contemporaries to be magical.49 Since
few scholars have any reservations about the authenticity of these two
traditions,50 it is highly likely that Jesus was written off by his opponents
as a magician, and thus as a social deviant.
Not surprisingly, already within both the Q tradition and the form of
the tradition known to Mark (as well as in the completed gospels) there
are responses to this charge. Followers of Jesus would not have transmitted
such a hostile assessment of him unless a firm refutation of the allegation
were juxtaposed.
The Q pericope contrasts sharply two ways of assessing the exorcising
activity of Jesus: is Jesus in league with the prince of demons, or are his
actions the result of his relationship to God? Jesus himself claims that his
exorcisms were carried out ‘by the finger of God’, as signs of the breaking
in of God’s kingly rule (Matt. 12.28 = Luke 11.20).
In later polemic the same issue arises, but without specific reference to
this passage. At I Apol. 30 Justin notes that critics claim that the miracles of
Jesus were done by magic arts, rather than as a result of Jesus’ relationship to
God as his Son. Celsus’ Jew argues similarly: on account of his miraculous
powers, Jesus gave himself the title God, but in fact they were the result of
magical powers gained in Egypt (Contra Celsum i.28). Celsus’ Jew closes his
remarks addressed to Jesus with the claim that the miracles were the actions
of one ‘hated by God and of a wicked sorcerer’ (Contra Celsum i.71); in
this way the claim of Christians that Jesus stands in a special relationship
to God is undermined.
As we have noted above, Mark himself has shaped considerably the
traditions he links together in 3.19b–35. In a redactional comment in verse
30 he points out that the saying of Jesus concerning blasphemy against the
Holy Spirit (3.28-9) applies to those who claimed that Jesus was possessed
by Beelzebul, and that he casts out demons by the prince of demons (3.22).
R. A. Guelich has perceptively summed up the key point Mark is making:
‘To attribute the work of the Spirit through Jesus to demonic forces is the
ultimate calumny for which there is no forgiveness.’51 Once again, as in the
49 See Aune, ‘Magic in Early Christianity’, pp. 1523–9; Smith, Jesus the Magician, (above, n. 22) pp. 94–
139.
50 These traditions critical of Jesus would have been an embarrassment to his followers, and so are most
unlikely to have been invented. For discussion of the criterion of embarrassment, see Stanton, The
Gospels and Jesus, pp. 174–7.
51 R. A. Guelich, Mark, Word Biblical Commentary (Dallas: Word, 1989), p. 180.
146 Jesus
Q Beelzebul traditions, there are two opposing assessments of the actions
of Jesus: are they to be ascribed to demonic possession or to divine agency?

It is more difficult to establish that in his own lifetime Jesus was considered
by some to be a false prophet who led Israel astray. In the gospels there is no
specific allegation along these lines. However, cumulative evidence makes
this a strong probability.
(1) As we have seen above, Matthew and John, and probably Luke,
were aware of this polemical charge. There is no reason to suppose that it
first arose at the time they wrote, for tension between Jews who accepted
Christian claims about Jesus and those who rejected them did not arise
overnight in the 80s.
(2) Given that Jesus was alleged by some to be a magician, and given the
close links between the ‘magician’ and ‘false prophet’ allegations to which
I drew attention above, it is highly likely that Jesus was said by some to be
a false prophet.
(3) John the Baptist was said to have a demon (Q: Matt. 11.18 = Luke
7.33). Since neither the synoptic evangelists nor Josephus (Antiquities 18.116-
19) attribute miracle-working powers to John,52 the polemical jibe recorded
in Q labels him as a demon-possessed false prophet. Since John and Jesus
were associated closely, Jesus was almost certainly also marginalized by some
with the same accusation.
(4) Although the evangelists do not emphasize that Jesus was a prophet,
in two sayings, Mark 6.4 and Luke 13.33, Jesus refers to himself as a prophet.
A number of other sayings and several of his actions confirm that he saw
himself as a prophet.53 Thus it would be surprising if some opponents
did not dub him a false prophet, perhaps even with Deuteronomy 13 in
mind.54

I have argued that the double allegation found in Justin’s Dialogue 69.7 and
in the rabbinic traditions quoted above has deep roots. In his own lifetime
Jesus was said by some to be a demon-possessed magician. It is probable,
but not certain, that he was also said to be a demon-possessed false prophet.
Several corollaries follow. It is generally accepted that in the first and
second centuries followers of Jesus and Jews who did not accept their claims
52 Cf. John 10.41. Mark 6.14 is an apparent exception, but on any view this is a puzzling verse.
53 I have summarized and discussed the evidence briefly in The Gospels and Jesus, pp. 194–9. See
especially D. E. Aune, Prophecy in Early Christianity and the Ancient Mediterranean World (Grand
Rapids: Eerdmans, 1983), pp. 153–89.
54 Strobel, Die Stunde der Wahrheit (above, n. 16), goes much further and argues that this accusation
was in fact central in the Sanhedrin trial.
 Jesus of Nazareth: a magician and a false prophet? 147
were at odds about Christology and the law. It is less frequently appreciated
that both the actions and the teaching of Jesus were also a source of tension
and dispute: they were assessed very differently by his later followers and
opponents.
The allegations of the contemporary opponents of Jesus confirm that he
was seen by many to be a disruptive threat to social and religious order. His
claims to act and speak on the basis of a special relationship to God were
rightly perceived to be radical. For some who heard the teaching and saw
the actions of Jesus, those claims were so radical and so unacceptable that
they had to be undermined by an alternative explanation of their source.
There is no reason to doubt that both the actions and the teaching of Jesus
brought him into conflict from time to time with the religious authorities
of the day. In that social setting, allegations of false prophecy and sorcery
thrive. As we noted above, they are used to marginalize and undermine the
influence of individuals whose claims and behaviour are perceived to pose
a threat to the stability of the dominant social order.
It is often urged that reconstructions of the life and teaching of Jesus must
account for his crucifixion on a Roman cross. If, for example, Jesus merely
echoed the conventional teaching and piety of his day, then it is difficult to
account for his downfall. In my judgement, the historian must also account
for the ‘aftermath’ of Jesus of Nazareth, i.e. for the stubborn fact that, in the
very conservative religious and social climate of the post-Easter decades,
followers of Jesus made profound claims about him and took radical steps
in his name. The claim advanced in this chapter that Jesus was perceived
in his own lifetime to be a demon-possessed magician / sorcerer and a false
prophet who deceived God’s people coheres well both with his downfall
and with the ‘aftermath’. Both the teaching and the actions of Jesus drew
criticism. Were his exorcisms and healing miracles signs that God’s kingly
rule was breaking into history, or were they the result of his collusion with
the prince of demons? Was Jesus a prophet sent by God, or was he a false
prophet who was deceiving Israel? The polemical traditions confirm that, as
Jesus moved around Galilean villages, his relationship to God was a central
issue.55

55 For fuller discussion of the points made in the final five paragraphs, see Stanton, The Gospels and
Jesus, pp. 292–9.
 chap t e r 7

Early objections to the resurrection of Jesus

Recent studies of early Christian traditions about the resurrection of Jesus

have generally overlooked the objections which were raised by opponents of
Christianity in the first and second centuries. This is somewhat surprising.
Historians know how important it is to consider evidence or arguments
which are an embarrassment to the eventual ‘winners’. Astute theologians
always listen carefully to the voices of ‘outsiders’.1
There are three reasons for taking this rather off-beat approach. (1) Criti-
cisms of early Christian claims concerning the resurrection of Jesus give us
some limited insights into the variety of ancient attitudes to life after death.
(2) They help us to appreciate more keenly the ways Christian proclama-
tion of the resurrection was understood or misunderstood by both Jews
and pagans. (3) By paying attention to early criticisms we may be able to
trace more readily the points at which early Christian traditions about the
resurrection have been shaped by apologetic concerns.
The potential value of this approach will be clear. So why have the voices
of the critics not been heard? They have been ignored mainly because it is
undeniably difficult to uncover the critics’ views. We have much more ex-
tensive evidence for early polemical comments on the actions and teaching
of Jesus.2 However, by casting the net widely I believe that it is possible
to make a number of observations which are relevant to inquiries into
the setting, development, and reception of early resurrection traditions,
observations which stimulate further theological reflection today.
Since it is often helpful in studies of earliest Christianity to work back
from later, clearer evidence to more problematic earlier evidence, I shall
start with objections to the resurrection of Jesus made in the second half of

1 This chapter is a revised version of my contribution to a Festschrift edited by Stephen Barton and
Graham Stanton, Resurrection (London: SPCK, 1994), pp. 79–94, in honour of my former colleague
at King’s College London, Leslie Houlden, who always goes out of his way to listen to those whose
viewpoints differ from his own.
2 See Chapter 6 above.

148
 Early objections to the resurrection of Jesus 149
the second century and work back, where appropriate, to New Testament
evidence. I am starting outside the canon for the further reason that it is all
too easy to use the New Testament texts as a mirror to find opponents and
rivals under every canonical bed. Evidence from outside the canon may
help us to avoid some of the pitfalls of mirror-reading.3

7.1 the pagan celsus and a jew

The first pagan comments on earliest Christianity come from the early
decades of the second century. Although Suetonius, Tacitus, Pliny,
Epictetus, Hadrian (in his imperial rescript), and Fronto provide invalu-
able insights into the ways Christians were perceived by pagans, they do
not help our quest for objections to the resurrection of Jesus. They tell us
that the odd behaviour of Christians raised many eyebrows, but they say
little or nothing about what Christians believed or taught.
In the second half of the second century, however, some pagan writers do
comment on Christian beliefs.4 Lucian of Samosata noted that Christians
associated with Peregrinus (a Christian convert who died about ad 165)
‘were poor souls who convinced themselves that they would all be immortal
and live forever, on account of which they think lightly of death’. Galen of
Pergamum also commented on the contempt Christians showed for death.
Lucian and Galen both remind us that belief in life after death (whether
resurrection or immortality) was by no means universal in the pagan world
of their day and that Christians were known for their distinctive hopes for
the future. However, neither writer refers explicitly to the resurrection of
Jesus.
For the first extended pagan objections to the resurrection of Jesus we
must turn to Celsus’ thorough-going attack on Christianity, which was was
written between 177 and 180. Some seventy years later, Origen recognized
that Celsus’ trenchant voice could be neither silenced nor ignored. So he
took the trouble to write a lengthy reply. Fortunately, Origen’s reply includes
quotations from as much as 70 per cent of Celsus’ now lost book.
Celsus’ wide-ranging attack includes comments on the resurrection of
Jesus, and on Christian beliefs in life after death. In the first sections of
his book the Platonist philosopher repeatedly quotes the views of a Jew. In

3 See John Barclay, ‘Mirror-Reading a Polemical Letter: Galatians as a Test Case’, JSNT 31 (1987) 73–93.
4 For the texts and comments, see S. Benko, ‘Pagan Criticism of Christianity during the First Two
Centuries’, ANRW ii.23.2, ed. W. Haase (Berlin: de Gruyter, 1980), pp. 1055–1118; R. L. Wilken, The
Christians as the Romans Saw them (New Haven and London: Yale University Press, 1984); and Molly
Whittaker, Jews and Christians: Graeco-Roman Views (Cambridge: Cambridge University Press, 1984).
150 Jesus
response, Origen insists that Celsus’ Jew is an ‘imaginary character . . . who
addresses childish remarks to Jesus’, and claims that the views attributed to
him are not consistent with those of a Jew (Contra Celsum i.28).5
However, there are good grounds for concluding that on this issue even
the learned Origen has nodded. Some of the critical comments attributed
to Celsus’ Jew concerning the birth, actions, and teaching of Jesus are also
found in polemical Jewish traditions and, independently, in other early
Christian writings.6 Celsus has clearly drawn on earlier Jewish traditions:
it is impossible to say just how early they are, but their value as evidence
for the views of Jewish opponents of Jesus and his followers can hardly be
overestimated.
In some passages it is not always possible to decide whether Celsus
is setting out his own objections to Christianity, or those of the Jewish
opponent whom he quotes. However, close attention to the text discloses
that Celsus and the Jew have quite different views on life after death and also
on the resurrection of Jesus. As in Luke–Acts – I shall discuss the evidence
below – it is possible to distinguish between pagan and Jewish objections.
Celsus claims that a dead man cannot be immortal (ii.16) and that
Christians worship a corpse (vii.68): for the pagan philosopher the very
notion of resurrection is absurd (v.14; vi.29), though he is willing to discuss
the possibility of some form of immortality (v.14; cf. iv.56). Celsus insists
that ‘the fact that hope of the resurrection of the dead is not shared by some
Jews [presumably Sadducees] and some Christians [presumably gnostics]
shows its utter repulsiveness, and that it is both revolting and impossible’
(v.14). However, he does not relate this observation to the resurrection
of Jesus. Hence it it not surprising to find that he does not discuss the
resurrection of Jesus at length: he leaves extended criticism to the Jew
whom he quotes.7 In contrast to Celsus himself, however, the Jew insists
that ‘Jews hope to be resurrected in the body and to have everlasting life’
(ii.77), a view which Origen accepts and expounds at some length (v.14-26).
At a later point in his attack on Christianity, long after he has stopped
quoting a Jew, Celsus returns to the evidence for the resurrection of Jesus
and mentions in passing the presence of women at the empty tomb, not one
hysterical woman, as in the objection of the Jew (cf. ii.55 and v.52). Rather
5 Quotations are from Henry Chadwick, Origen: Contra Celsum (Cambridge: Cambridge University
Press, 1953).
6 See G. N. Stanton, A Gospel for a New People: Studies in Matthew (Edinburgh: T. & T. Clark, 1992),
pp. 171–2, 185–9.
7 See E. Bammel, ‘Der Jude des Celsus’, in his collected essays, Judaica: Kleine Schriften I (Tübingen:
Mohr, 1986), pp. 265–83. Bammel concludes that, since the comments of Celsus’ Jew come from the
time of the apostolic fathers and the later writers of the NT, they are of unsurpassed value.
 Early objections to the resurrection of Jesus 151
surprisingly, Celsus himself makes nothing of the fact that Christian claims
rest on the evidence of women. Instead he points out scornfully that the
alleged Son of God ‘was not able to open the tomb, but needed someone
else to move the stone’ (v.52). This is an objection which Celsus’ Jew is
unlikely to have made; in a Jewish context, ‘Son of God’ did not necessarily
denote divinity.
In short, the pagan Celsus and his Jewish ally raise rather different objec-
tions to the resurrection of Jesus. Celsus uses the Jew’s specific objections
to supplement his own scornful comments on the whole notion of post-
mortem resurrection.

Pagan parallels to the resurrection of Jesus are discussed in a particularly

interesting passage in ii.55-8. At this point Celsus’ Jew is addressing Jews
who believe in Jesus. He claims that six Greek legends provide parallels
to the resurrection of Jesus. ‘Do you think that the stories of these others
really are the legends which they appear to be, and yet that the ending of
your tragedy is to be regarded as noble and convincing?’ The implication is
that Christian Jews believe that the resurrection of Jesus cannot be put on
a par with other stories about individuals brought back to life. But nothing
is made of this.
Origen’s own defence is striking: whereas the heroes in the Greek le-
gends disappeared secretly and then returned to the men they had left,
‘Jesus was crucified before all the Jews and his body put to death in the
sight of their people.’ However, he then goes on to claim that there are
important Biblical parallels: the raising to life of the young boys by Elijah
(I Kgs 17.21-2) and Elisha (II Kgs 4.34-5). The resurrection of Jesus, he
insists, was more remarkable because of the greater benefits it brought to
mankind.
No doubt this latter comment is valid, but Origen misses a far more
important point: the young boys to whom he refers, along with Jairus’
daughter, the widow’s son at Nain, and Lazarus, were restored to this life.
The evangelists, for all their differences in emphasis, did not claim that the
body of Jesus was resuscitated only to die again at a later point. Perhaps in
their refusal to appeal to parallels to the resurrection of Jesus, the Christian
Jews addressed by Celsus’ Jew saw this more clearly than either Origen or
Celsus. One hopes so.
Celsus’ Jew advances vigorously the theory that the followers of Jesus
‘saw’ their recently crucified leader in a dream or hallucination. Origen’s
response is not very persuasive: ‘Celsus’s idea of a vision in the daytime is not
convincing when the people were in no way mentally disturbed and were
152 Jesus
not suffering from delirium or melancholy. Because Celsus foresaw this
objection he said that the woman was hysterical; but there is no evidence
of this in the scriptural account’ (ii.60).
This discussion reminds us of the ultimate futility of trying to seek proof
one way or the other. ‘Vision’ or ‘hallucination’, how can one decide? Surely
the matter can be settled only on the basis of wider considerations, which
are theological rather than historical or psychological.
Three of Celsus’ objections elicit a similar theological response from
Origen. Celsus claims that, if Jesus really wanted to show forth divine
power, he ought to have appeared to those who maltreated him, to the man
who condemned him, and ‘to everyone everywhere’ (ii.63, 67). In reply to
this threefold challenge, Origen insists that, when Jesus was sent into the
world, he did not merely make himself known; he also concealed himself
(ii.67). Similarly (with reference to the baptism as well as the resurrection
appearances), ‘the divine voice is such that it is heard only by those whom
the speaker wishes to hear it’ (ii.72). Perhaps the sharpest objection raised
by Celsus’ Jew is his jibe, ‘Where is the risen Jesus, that we may see and
believe?’ In response, Origen turns the tables on his Jewish opponent and
asks for proof that Israel is God’s ‘portion’ (ii.77). Origen’s riposte is terse
and to the point: God’s self-disclosure to Israel was not openly accessible
to all and sundry on a permanent basis, and its reality cannot be proven.
So too with the appearances of the risen Jesus.
A theological thread runs through these responses to Celsus that is as
important for current theological reflection as it was in the middle of the
third century. Origen sees clearly that, although some of Celsus’ objections
were based on misunderstandings and others were mischievous, he himself
could not offer clear-cut historical proof of the resurrection of Jesus (see
especially ii.77): God is not at the beck and call of men and women.
Origen knows full well that proof of the historicity of an incident in
the gospels is difficult and in some cases impossible (i.42). He knows that
he cannot sidestep allegations that the text of the gospels has been tam-
pered with (ii.27) and that the resurrection narratives contain discrepancies
(v.55-6). He repudiates ‘mere irrational faith’, and insists that readers of the
gospels need an open mind and considerable study. ‘If I may say so’, he
writes, ‘readers need to enter into the mind of the writers to find out with
what spiritual meaning each event was recorded.’ Is this the way faith and
reason should be held together in discussion of the resurrection narratives?
Is there still a place for discerning ‘spiritual meaning’ by Origen’s own
method of allegorical interpretation? If so, what criteria will guard against
‘irrational faith’?
 Early objections to the resurrection of Jesus 153
Nearly all the objections to the resurrection of Jesus which have been
raised since the Enlightenment were anticipated by the pagan Celsus, or by
the Jew whom he quotes. There is only one notable exception, the theory
that the disciples stole the body of Jesus from the tomb.8 Most of the
objections to the resurrection of Jesus which are pressed by Celsus’ Jew are
likely to have been made long before Celsus wrote, between ad 166 and
170. Their importance for serious discussion of New Testament resurrection
traditions cannot be overestimated. Some of Origen’s replies carry little or
no persuasive power today, but some are (or should be) still on the agenda
of current theological discussion.

7.2 trypho the jew

Just over a decade before Celsus launched his attack on Christianity, Justin
Martyr wrote his Dialogue with the Jew Trypho. Four passages are of par-
ticular interest.
(1) Justin and Trypho, just like Origen and Celsus (and, as we shall see
shortly, Luke’s Paul and his Jewish opponents), can agree that there will be
a resurrection of the righteous (45.2-3).
(2) Justin, just like Celsus, knows that there are some so-called Christians
(presumably gnostics) who say that there is no resurrection of the dead. He
is confident that Trypho will agree that Sadducees (and some other Jewish
groups) who deny the resurrection should not be considered true Jews
(80.3-4). We shall see below that Luke also sees hope for resurrection from
the dead as a line of continuity between Judaism and Christianity.
(3) In one passage a Christian credal summary is put into the mouth
of Trypho. ‘You say many blasphemous things’, says Trypho, ‘thinking to
persuade us that this man who was crucified has been with Moses and
Aaron, and has spoken to them in a pillar of a cloud, that he then became
man and was crucified, and has ascended into heaven, and comes again
on earth, and is to be worshipped’ (Dialogue 38.1). Just as in several New
Testament passages (e.g. Phil. 2.5-11; I Tim. 3.16; Eph. 4.7-10; Rom. 10.5-8),
this ‘creed’ moves directly from the crucifixion to the exaltation of Jesus
without referring to resurrection.
(4) I noted above that the only major objection to the resurrection of
Jesus which is not mentioned by Celsus is the claim that the disciples stole
8 In i.51 Origen mentions, but only in passing and not in response to an objection raised by Celsus,
Matthew’s attempt to rule out the claim that the disciples stole the body of Jesus (28.13-14). Origen
even claims that the soldiers who guarded the tomb and were later bribed were eyewitnesses of the
resurrection, as in the Gospel of Peter 9–10!
154 Jesus
the body of Jesus. By a curious irony, this is the only Jewish criticism of
Christian claims about the resurrection which Justin does mention. Justin
claims that on hearing that Christ had risen from the dead, Jewish leaders
appointed chosen men and sent them into all the civilized world, proclaiming that
‘a certain godless and lawless sect has been raised by one Jesus of Galilee, a deceiver,
whom we crucified, but his disciples stole him by night from the tomb, where he
had been laid after being unnailed from the cross, and they deceive men, saying
that he is risen from the dead and has ascended into heaven’. (Dialogue 108)9
Some of the phraseology of this alleged anti-Christian Jewish propa-
ganda comes from Justin himself, some of it possibly from Matt. 28.13,
15. However, there are good reasons for supposing that Justin may here be
drawing on an earlier source, for these Jewish allegations and not simply
on New Testament passages. (1) At this point at least, Justin is not setting
up Trypho as a straw man who lists Jewish objections in order to allow
Justin to refute them one by one, for Justin does not respond to them
anywhere in the Dialogue. (2) The reference to Christianity as a ‘sect’ is
striking and unique in Justin: he uses the term elsewhere to refer to factions
within Judaism (62.3; 80.4) and within Christianity (35.3; 51.2), but not to
Christianity per se. (3) The reference to Christianity as a godless and lawless
sect raised by Jesus does not come from the gospels, nor does the reference
to the disciples as deceivers.
While it is impossible to confirm that a Jewish counter-mission of this
kind took place, there is nothing inherently improbable in the reference
to Jewish messengers being sent ‘into all the civilized world’.10 Justin is
probably drawing on (as well as filling out himself ) early Jewish allegations.
Certainly Dialogue 108 sets out what Justin felt to be the heart of Jewish
arguments against the resurrection of Jesus. This chapter confirms that
Christians were aware of, and sensitive to, Jewish claims that the disciples
stole the body of Jesus from the tomb. But, unlike the evangelist Matthew,
Justin does not bother to refute this obvious objection to Christian claims.
Perhaps he felt that it was so absurd that refutation was unnecessary.
In some versions of the Jewish polemical traditions known as the Toledoth
Jeshu, there is an interesting variant on the theory that the body of Jesus

9 Justin refers to this ‘counter-mission’ near the opening of the Dialogue (ch. 17), but at that point he
gives only a summary of its contents, without mentioning the objection to the resurrection; there is
an even briefer reference in ch. 117.
10 Following formal declaration in Jerusalem of observation of the new moon, messengers were sent
to the diaspora. My former research student Dr Eileen Poh has drawn my attention to references
to letters on other matters which were sent to the diaspora: Jer. 29. 4-23; 2 Baruch 78-87; 2 Macc.
1.1-10a; 10b–2.18; Esther 9.20-32.
 Early objections to the resurrection of Jesus 155
was stolen from the tomb. The body of Jesus is stolen, not by his disciples
as in Matthew and in Justin, but by his opponents: they drag the body
through the streets of Tiberias.11 These traditions almost certainly stem
from a time after the conversion of the Emperor Constantine: by then Jews
and Christians were no longer rivals, and Jews were all too aware of the
heavy hand of Christian oppression. This form of anti-Christian polemic
falsifies Christian claims at a stroke; there is no need for the elaborate ruses
Matthew alleges were adopted by ‘the Pharisees’ and by ‘the chief priests
and elders’ (27.62-6; 28.11-15).
The fact that the passage quoted above from Dialogue 108 is the only
reference to Jewish objections to the resurrection of Jesus and that there
is no debate between Justin and Trypho on this topic is striking. Why
do Justin and Trypho share convictions about a general resurrection of
the dead (Dialogue 45.2-3), but fail to discuss the resurrection of Jesus?
For them both, Christology and the law are more important issues. Their
central dispute concerning Jesus of Nazareth is not his resurrection, but
whether or not Jesus fulfils Scripture; as we shall see in a moment, this was
also Luke’s view.
As in other Jewish polemic from this period, the claim that Jesus was
a magician and a deceiver who led Israel astray is much more prominent
than objections to the resurrection of Jesus. Perhaps the latter was not
an easy issue to debate. After all, many Jews accepted that one day God
would raise the (righteous) dead: how could one decide whether or not this
general resurrection had been anticipated in the case of Jesus? As in Justin’s
Dialogue, debate could be neatly sidestepped by the claim that the disciples
had stolen the body of Jesus.

7.3 g entile and jewish responses in acts

I turn now from Justin to an earlier Christian apologist. Luke is aware of
rather different negative reactions to Christian proclamation of the resur-
rection of Jesus. I shall start with his presentation of Gentile objections to
the resurrection of Jesus.
Paul ends his speech to the ‘Athenians’ in the Council of the Areopagus
with the claim that God has raised from the dead the man whom he has
designated to judge the world (Acts 17.31). Paul’s listeners do not comment
on his proclamation of God the Creator who does not live in shrines made

11 See E. Bammel, ‘The Titulus’, in Ernst Bammel and C. F. D. Moule, Jesus and the Politics of his Day
(Cambridge: Cambridge University Press, 1984), pp. 361–2.
156 Jesus
by human hands, but some of them scoff at his reference to resurrection
from the dead (17.32). From Luke’s point of view, Gentiles find the notion
of resurrection quite incomprehensible.
Here Luke is echoing widely held views. The prevailing Greek attitude
towards resurrection is summed up by Aeschylus, Eumenides 647-8. On
the occasion of the founding of the court of the Areopagus in Athens, the
god Apollo observes, ‘Once a person is slain and the dust has drunk up his
blood, there is no resurrection (ˆn†stasiv)’.
There may be a further reference in Acts 17.18 to a negative response in
Athens to Paul’s proclamation of the resurrection, but I am not as confident
about this as most translators and commentators. Luke opens his account
of Paul’s visit to Athens by noting that he argued in the marketplace every
day with passers-by, some of whom were Epicurean and Stoic philosophers.
As is often the case in Acts, there is a division of opinion. Some said, ‘What
does this babbler want to say?’ Others said, ‘He seems to be a proclaimer
of foreign divinities.’ In most editions of the Greek text and in modern
translations Luke then explains in an aside to the reader that these critical
comments were responses to Paul’s proclamation of ‘the good news about
Jesus and the resurrection’ (17.18).
Many scholars now accept that in this lengthy verse Luke claims that
the Athenians totally misunderstood Paul’s proclamation of Jesus and the
resurrection (anastasis) as a reference to two divinities, ‘Jesus and Anastasis’.
This ingenious explanation has an impressive pedigree, which goes back
to Chrysostom; it is supported by the NEB’s translation in 17.18, ‘Jesus
and Resurrection’.12 But the explanation is unlikely, for it implies in an
un-Lucan way either that Paul was a completely ineffective communicator,
or that his listeners were stupid. In addition, in Codex Bezae (supported
by the old Latin codex Gigas) there is important, though often overlooked,
evidence for omission of the whole aside at the end of verse 18: a very
early copyist may have noted Luke’s failure to indicate the content of Paul’s
initial proclamation to the Athenians, and inserted the aside on the as-
sumption (based on 17.32) that Paul must have preached about ‘Jesus and
the resurrection’ and been totally misunderstood.13
12 REB is a little less confident; ‘Jesus and the Resurrection’. NRSV and NIV have ‘Jesus and the
resurrection’.
13 Since Codex Bezae (and the so-called western text) includes a large number of additional words and
phrases in Acts (often as explanatory comments), in some of the small number of places where this
codex offers a shorter reading it may well retain the original wording. It is less easy to account for
later omission of the phrase than for a later addition. W. A. Strange, The Problem of the Text of Acts
(Cambridge: Cambridge University Press, 1992), pp. 188–9, suggests that addition of the explanatory
clause at the end of Acts 17.18 may have been the clarification of a copyist who was concerned that
 Early objections to the resurrection of Jesus 157
In the second passage in Acts which refers to Gentile reaction to Christian
proclamation of the resurrection, the Roman governor Festus informs
Agrippa that he had been surprised by the accusations the chief priests
and elders of the Jews had brought against Paul. ‘They merely had certain
points of disagreement with him about their religion, and about someone
called Jesus, a dead man whom Paul alleged to be alive’ (Acts 25.19, REB).
Festus quickly concedes that he is out of his depth in such discussions.
Luke then draws a careful contrast between Gentile and Jewish responses
to proclamation of the resurrection. When Paul defends himself before
Agrippa, he assumes that the Jewish king will not find it incredible that God
should raise the dead (26.8). He then claims that in his preaching he asserts
nothing beyond what was foretold by the prophets and by Moses: ‘that the
Messiah would suffer and that, as the first to rise from the dead, he would
proclaim light both to the Jewish people and to the Gentiles’ (26.22-3). At
mention of the resurrection of the Messiah, it is not the Jewish king Agrippa
but the Roman governor Festus who intervenes: ‘Paul, you are raving; too
much study is driving you mad.’ Paul does not attempt to enlighten Festus
on the subject of the Messiahship of Jesus and the resurrection. Instead, he
draws Agrippa into the discussion and repeats his claim that, since Agrippa
is a Jew, he is well versed in these matters (26.24-6).
This passage is the final reference to the resurrection in Acts. In it Luke
sets out clearly his belief that, while Gentiles are completely baffled by
Christian proclamation of the resurrection of Jesus, at least some Jews
responded (or should have responded) much more sympathetically.
This latter point is developed by Luke in a rather subtle way which is
well worth exploring, for it raises important theological issues. Luke first
isolates Sadducean denial of the resurrection of the dead as a minority Jewish
opinion (Acts 4.1-4; 5.17; 23.6). Luke believes that other Jews, including
the Pharisees, do not have a fundamental objection to Christian claims
concerning the resurrection of the dead. Hence, so the argument runs,
they ought not to dismiss Christian claims concerning the resurrection of
Jesus.
In the account of the trial of Paul by the Sanhedrin (Acts 23.1-10), Paul
emphasizes that he is himself a Pharisee, a son of Pharisees, and then singles
out ‘the hope and resurrection of the dead’ as the central issue at stake.
The Sanhedrin is divided. Luke reminds the reader that the Sadducees
say that there is no resurrection and then notes that ‘some scribes of the

the reference to Paul preaching strange daimon©a should not become the growth point for Gnostic
speculation.
158 Jesus
Pharisees’ defend Paul: ‘we find nothing wrong with this man’ (23.9). They
even concede the possibility that the appearance of Christ to Paul outside
Damascus was a reality: a spirit or angel has spoken to him.14
A similar point is made in Paul’s defence before Agrippa (26.2-23). Luke’s
Paul refers once again to his membership of the ‘sect’ of the Pharisees. An
important switch from the singular to the plural in verse 8 (missed by
many translators) indicates that Paul is addressing not merely Agrippa,
but Jews generally: they should not think it incredible that God raises the
dead! From the context it is clear that the resurrection of Jesus is in view.
The line of argument is bold, to say the least. Luke’s Paul is claiming that,
on the basis of Jewish (especially Pharisaic) beliefs about the resurrection
from the dead, the resurrection of Jesus is not at all problematic. In Luke’s
view, ‘resurrection’ is one of the lines of continuity between Judaism and
Christianity.15
This reading is consistent with the earlier chapters of Acts. In numerous
passages Luke records the objections Jews raised to Stephen’s and Paul’s
preaching. Although the resurrection of Jesus is prominent in the preceding
‘sermons’ (or, better, ‘speeches’), with the exception of Sadducees and their
allies, objections are not raised by Jewish opponents.
However, in two passages Luke makes a key theological point, which he
could hardly have expected non-Christian Jews to accept: the resurrection
of Jesus is the first example of the resurrection of the dead. In rather terse
Greek at Acts 4.2, Luke insists that the resurrection had come to pass in
the person of Jesus.16 He repeats the point in his final reference to the
resurrection: Paul proclaims to Agrippa ‘that the Messiah must suffer, and
that, by being the first (prätov) to rise from the dead’, he would proclaim
light ‘both to our people and to the Gentiles’ (26.23). Luke believes that as
a Jew Agrippa will not find the notion of resurrection from the dead absurd
(26.8). However, the claim that ‘Jesus is the first to rise from the dead’ must
be asserted vigorously, for it will not find a ready welcome.

14 So E. Haenchen, The Acts of the Apostles (Oxford: Blackwell, 1971), pp. 638–9.
15 Cf. ibid., p. 659: ‘The new faith . . . is not a treason to the old. The hope of resurrection is the bond
which holds the two together. That the Sadducees do not believe in the resurrection means only that
it is a question within the Jewish faith in which Rome need not interfere.’
16 So also ibid., p. 214. C. K. Barrett is not persuaded: ‘That the resurrection of Jesus was the firstfruits
of the final resurrection is a NT thought, but it is not clear that Luke entertained it.’ Acts, Vol. i, ICC
(Edinburgh: T. & T. Clark, 1994), p. 220. However, in his comments on 4.2, Barrett fails to note
26.23. In his comments on the latter verse he allows that the interpretation offered above is plausible.
NRSV echoes the terse Greek: ‘they were proclaiming that in Jesus there is the resurrection of the
dead’. REB attempts to unpack the Greek and almost misses the point: ‘they were proclaiming the
resurrection from the dead by teaching the people about Jesus’.
 Early objections to the resurrection of Jesus 159
Luke sets the resurrection of Jesus into an eschatological framework
which is distinctively Christian. In addition to the two passages just men-
tioned, the four passages which refer to the hope of Israel make this clear.
In 23.6 (‘the hope and the resurrection of the dead’); 24.15f.; 26.6f.; and
28.20 Luke’s Paul insists that the resurrection of Jesus should be seen as the
climax of the messianic hopes of Israel.17
For Luke the resurrection of Jesus is the fulfilment ‘of what the prophets
and Moses said would take place’ (Acts 26.23). The same points had al-
ready been made strongly in Luke’s Gospel: 24.21, 26-7, 32, 44-6. Perhaps
somewhat optimistically, Luke believes that most Jews will not have fun-
damental objections to the resurrection of Jesus. What separates most Jews
and most Christians from one another is not the historical evidence for the
resurrection of Jesus, but whether or not Jesus fulfils the messianic hopes
of Israel. To the surprise of many Christians, the contemporary Jewish the-
ologian Pinchas Lapide defends a similar position. He openly accepts the
resurrection of Jesus as a historical event and as an act of God. But for
Lapide, as for Luke, it is the Messiahship of Jesus which marks the division
between Christianity and Judaism.18

7.4 other first-century object ions

In the light of the preceding discussion, several further New Testament
passages need to be reconsidered, though in each case the dangers of mirror-
reading, to which I drew attention above, should be borne in mind. I noted
above that Celsus presses the objection that, if Jesus had really wanted to
show forth divine power, he ought to have appeared to those who put him
to death, and to everyone everywhere (ii.63, 67). Perhaps similar objections
lie behind Acts 10.40-2. Luke’s ‘answer’ to the objection that the risen Jesus
was not seen by the whole people is striking: the chosen witnesses to whom
Jesus had appeared were commanded to proclaim him to the people.
John 14.18-22 may also reflect similar consternation over the resurrection
appearances. In these verses it is not clear whether resurrection appearances
or the parousia (or both) are in view. In verse 22, however, the question
of the ‘other’ Judas certainly refers to the resurrection appearances: ‘how is
it that that you intend to disclose yourself to us, and not to the world?’.
From the immediate context, John’s answer to this possible objection is that

17 On these passages, see K. Haacker, ‘Das Bekenntnis des Paulus zur Hoffnung Israels nach der
Apostelgeschichte des Lukas’, NTS 31 (1985) 437–51.
18 Pinchas Lapide, The Resurrection of Jesus (London: SPCK, 1984).
160 Jesus
Jesus will disclose himself to those who love him and keep his commands
(verse 21).19
In Mark 15.44-5 we read that ‘Pilate wondered if he were already dead;
and, summoning the centurion, he asked him whether he had been dead
for some time. When he learned from the centurion that he was dead, he
granted the body to Joseph.’ These verses may also be related to an early
objection to the resurrection of Jesus. Although Matthew and Luke follow
Mark closely in this pericope, these two verses are not found in either gospel.
So I think it is probable that these two verses were added very early to the
version of Mark used by Matthew and by Luke. They were intended to
underline the reality of the death of Jesus and so rule out a claim that Jesus
had not really died on the cross, but had revived in the cool tomb. If this
explanation is accepted, the ‘swoon’ theory, like most modern objections
to the resurrection of Jesus, may have been anticipated in antiquity.
By now some readers of this chapter will have recalled the objection
mentioned by Paul at I Cor. 15.12: ‘If . . . we proclaim that Christ was raised
from the dead, how can some of you say there is no resurrection of the
dead?’ I do not think that this verse reflects any doubt on the part of
the Corinthians about the resurrection of Jesus: there is no hint either in
the immediate context or elsewhere in I Corinthians that this was the case.
For reasons which we need not discuss here, some Christians in Corinth
denied that believers could or should look forward to a general resurrection.

The early objections to the resurrection of Jesus remind us of the varied

views of both Jews and pagans on post-mortem existence. It is not very
surprising that some of the New Testament resurrection traditions have
been shaped in the light of a range of apologetic concerns. More surprising,
however, are the ways in which the evangelists and those who earlier trans-
mitted resurrection traditions offered hostages to fortune. In most of the
traditions, discovery of the empty tomb is bewildering and does not lead to
faith. The risen Jesus appears only to a small number of his followers – and
their doubts are not excised from the traditions. Several obvious objections
are not countered: resurrection traditions are transmitted and proclaimed,
doubts and all.
Early objections to the resurrection hardly ever seem to have been
made in isolation from negative assessments of the teaching and the ac-
tions of Jesus.20 Opponents and followers alike saw that claims about the
19 I am grateful to the Revd Roger Larkinson for pointing out to me the possible relevance of this
passage for my approach.
20 For the latter, see Chapter 6 above.
 Early objections to the resurrection of Jesus 161
resurrection of Jesus raised the same issues as his actions and his teaching:
for opponents, the whole story was riddled with trickery and deceit; for
followers the story was God’s story.
An even more important issue emerges from this study of early objections
to the resurrection of Jesus. For some early opponents of Christianity, the
theological claims about the Risen Jesus which were made by his followers
raised far more fundamental issues than did allegedly implausible details in
the resurrection traditions. The opponents who saw this still deserve to be
listened to.
 part iii

The gospels and papyri codices

 chap t e r 8

Why were early Christians addicted to the codex?

The transition from the roll format to the codex-book format in the early
centuries of the Christian era was at least as revolutionary as its two later
counterparts. Who doubts the importance of Johannes Gutenberg’s mid-
fifteenth-century invention of printing by using movable type? Or the late
twentieth-century emergence of a CD-ROM only 12 cm in diameter yet
able to hold a very large book in digital form?
In this chapter I shall refer to new evidence and try to account for early
Christian ‘addiction’ to the codex along partly fresh lines. But I cannot
hope to solve all the problems, for there are still too many gaps in our
knowledge. I shall argue that attention to the origin of the codex forces
us to reconsider several important issues. Even before Paul wrote his first
‘canonical’ letter c. ad 50, followers of Jesus were accustomed to use the
predecessors of the codex-book format, various kinds of ‘notebooks’. They
used them for Scriptural excerpts and testimonies, for drafts and copies of
letters, and probably also for collections of traditions of both the actions
and the teaching of Jesus.
I owe the phrase ‘addiction to the codex’ to Sir Frederic G. Kenyon.
His publication in 1933–7 of the Chester Beatty Biblical papyri sparked off
interest in the emergence of the codex which continues to this day, for all
twelve Chester Beatty manuscripts are in the codex format. For a decade or
so the Chester Beatty manuscripts retained their standing among Biblical
scholars as the sensation of the age. In the following decades, however, they
were overshadowed somewhat by the even more dramatic discoveries of
the Dead Sea Scrolls and the Nag Hammadi Library in the 1940s, and the
Bodmer papyri in the 1950s.
The Chester Beatty papyri were published remarkably quickly by Kenyon
in handsome editions which remain invaluable for specialists seventy years

165
166 The gospels and papyri codices
later.1 Kenyon’s comments on the collection as a whole and on the individual
codices are astute, though he himself did not appreciate fully the importance
of some of his observations.2 Kenyon noted that the twelve Chester Beatty
manuscripts revolutionize our knowledge of the early history of the codex
form of book. He observed that the earlier part of the fourth century
had been taken as the date of the supersession of the roll by the codex.
The Chester Beatty papyri, however, not only ‘confirm the belief that the
Christian community was addicted to the codex rather than to the roll, but
they carry back the use of the codex to an earlier date than there has hitherto
been any good ground to assign to it’.3 The Chester Beatty papyri seemed
to confirm use of the codex in the second century, ‘and even probably in
the earlier part of it’.4
Kenyon noted that, so long as the roll was the form of book in use, no
work of materially greater length than one of the gospels could be contained
in a single roll. So, whereas five rolls would have been needed for the four
gospels and Acts, among the Chester Beatty papyri one codex from the
third century contained all these writings.5 Kenyon was well aware of the
significance of this new evidence for the physical existence of the fourfold
Gospel. However, he seems not to have noted that this codex contains rare
and important evidence for the joint circulation of Luke’s Gospel and Acts,
albeit almost certainly separated in this codex by Mark’s Gospel.6
In at least one respect Kenyon’s comments on the significance of the
Chester Beatty papyri were truly prophetic. ‘When, therefore, Irenaeus at
the end of the second century writes of the four Gospels as the divinely
provided evidence of Christianity, and the number four as almost axiomatic,
it is now possible to believe that he may have been accustomed to the sight
of volumes in which all four [Gospels] were contained.’7 In 1933 those

1 F. G. Kenyon, The Chester Beatty Biblical Papyri: Descriptions and Texts of Twelve Manuscripts on
Papyrus of the Greek Bible (London: Emery Walker, 1933–7).
2 Kenyon’s textbooks were published in several editions and were widely influential. But neither he
nor the later editors of his textbooks (Kenyon died in 1949) developed his comments on the origin of
the codex format; hence his own contribution has not received the recognition it deserves. See, for
example, The Text of the Greek Bible, 3rd edn, revised by A. W. Adams (London: Duckworth, 1975),
p. 9, and Our Bible and the Ancient Manuscripts (London: Eyre and Spottiswoode), pp. 41–3.
3 Kenyon, Chester Beatty, Vol. i, p. 12. The italics are my own.
4 This claim was based on Kenyon’s proposed dating of the Numbers and Deuteronomy codex: ‘written
in a fine hand, of the second century, perhaps of the first half of it’ (Vol. i, p. 8). Although A. S.
Hunt’s more cautious late second- or early third-century dating has won the day, no one now doubts
that Christians used the codex in the first half of the second century, if not even earlier.
5 Kenyon, Chester Beatty, Vol. i, p. 12.
6 See T. C. Skeat, ‘A Codicological Analysis of the Chester Beatty Papyrus Codex of Gospels and Acts
(P45)’, Hermathena 155 (1993) 27–43.
7 Kenyon, Chester Beatty, Vol. i, p. 13.
 Why were early Christians addicted to the codex? 167
comments must have seemed to many Biblical scholars to be unduly spec-
ulative. For many decades little notice was taken of them. However, more
recent discoveries have enabled this insight to be placed on a much firmer
footing, for we now have almost certain evidence for the existence of two
earlier codices which originally contained all four canonical gospels.8 The
theological case for acceptance of the fourfold Gospel and the availability
and pragmatic advantages of a single codex for four gospel writings did not
develop independently. Canon and codex are indeed inter-related.

8 .1 ‘big bang’ theories

Sir Frederic Kenyon did not discuss the reasons for early Christian ad-
diction to the codex. Twenty years later, in 1954, C. H. Roberts took up
the challenge in ‘The Codex’, the published form of two lectures, which
remains a landmark in modern discussion.9 Roberts suggested that it was
Mark’s Gospel which provided the inspiration for the codex. The gospel,
he claimed, was written as a parchment notebook in Rome. When it (soon)
reached Egypt, it is likely that it would have been copied on papyrus, so
much more readily available there than parchment, and the papyrus codex
might thus have been created.
Roberts proposed a ‘big bang’ theory to account for the development
of the codex and for early Christian addiction to this format: the use of
the codex for Mark’s Gospel by the church in Alexandria was the trigger
factor for early Christian use of the codex. In 1983 a completely revised and
extended version of ‘The Codex’ was published as a monograph by C. H.
Roberts and T. C. Skeat with the title The Birth of the Codex.10 In spite
of more recent discoveries, this learned book remains the starting point
for serious discussion of our question: its presentation and assessment of
the Latin literary evidence and its statistics for the codex and roll in the
early Christian centuries are invaluable. However, in one important respect
the authors’ approach was misguided, for they continued to search for one
major trigger factor to account for the birth of the codex.
Roberts and Skeat gave reasons for setting aside Roberts’s own 1954
‘big bang’ theory and then proposed an alternative.11 They drew on S.
Liebermann’s claim that rabbinic sayings were written down on papyrus

8 For full discussion, see Chapter 3 above.

9 C. H. Roberts, ‘The Codex’, Proceedings of the British Academy 40 (1954) 169–204.
10 C. H. Roberts and T. C. Skeat, The Birth of the Codex (London: Oxford University Press for the
British Academy, 1983).
11 Ibid., pp. 54–61.
168 The gospels and papyri codices
tablets (p©nakev), and proposed that sayings of Jesus may have been written
down on papyrus tablets which were then developed into a primitive form
of codex in Antioch, ‘a Proto-Gospel’.12
In 1994 T. C. Skeat returned to the question yet again.13 He noted that
neither of the hypotheses advanced in The Birth of the Codex had won accep-
tance, and so he approached the problem from a different standpoint. ‘We
must assume’, Skeat wrote, ‘in the absence of any evidence to the contrary,
that the Gospels originally circulated in the usual way, on papyrus rolls.
What can have induced the Church so suddenly and totally to abandon
rolls, and substitute not just codices but a single codex containing all four
Gospels?’14 Skeat’s answer to his question is that the publication of John’s
Gospel (c. ad 100) may well have caused a crisis in the church. Believ-
ers must have been in a quandary: when would the production of gospels
come to an end? So there must have been correspondence between the ma-
jor churches, and perhaps conferences. The decision was reached to include
all four Gospels in a codex, ‘the new form of book recently developed in
Rome’ – and this would have been quite impossible in the universally used
roll format. ‘And once the Four-Gospel codex had been decided upon,
every means must have been taken to spread the news throughout the
Church.’15
Skeat’s 1994 theory raises more questions than it answers. It is yet another
‘big bang’ theory, for it proposes one trigger event to account for the early
church’s initial acceptance of the codex format and its rapid addiction.
In 1996 David Trobisch published a partly similar theory: early Christian
use of the codex and of nomina sacra (the distinctive Christian method of
abbreviation of ‘sacred names’ such as ‘God’ and ‘Jesus’) can be accounted
for only by positing deliberate decisions concerning the appropriate format
for canonical writings.16
In 1990 Harry Gamble published a very different ‘big bang’ theory.17
Gamble accepted that ‘there must have been a decisive, precedent-setting
development in the publication and circulation of early Christian literature
that rapidly established the codex in Christian use, and it is likely that this

12 This hypothesis is discussed further below, pp. 186–8.

13 T. C. Skeat, ‘The Origin of the Christian Codex’, ZPE 102 (1994) 263–8.
14 Ibid., 266. 15 Ibid., 267.
16 D. Trobisch, Die Endredaktion des Neuen Testaments: Eine Untersuchung zur Entstehung der christlichen
Bibel (Freiburg:Universitätsverlag; Göttingen: Vandenhoeck & Ruprecht, 1996).
17 H. Y. Gamble, ‘The Pauline Corpus and the Early Christian Book’, in W. Babcock, ed., Paul and
the Legacies of Paul (Dallas: Southern Methodist University Press, 1990), pp. 265–80. A summary of
the theory was included in his Books and Readers in the Early Church: A History of Early Christian
Texts (New Haven and London: Yale University Press, 1995), pp. 58–65.
 Why were early Christians addicted to the codex? 169
development had to do with the religious authority accorded to whatever
Christian document(s) first came to be known in codex form. There are
good reasons to think that this distinction belonged to an early edition of
the letters of Paul.’18
So there are at present a handful of ‘big bang’ theories ‘on the table’.
Alas, none of them is persuasive. Both Skeat and Trobisch envisage a much
higher level of structure, centralized organization, and decision-making
procedures within early second-century Christianity than I think likely. I
do not think that we can posit the emergence of the four-gospel codex early
enough to have been the trigger factor.19 And Gamble has not convinced
me that the collection of Paul’s letters into a single codex is likely to have
persuaded all strands of early Christianity to adopt the new codex format
not only for Paul’s writings, but also for gospels. E. J. Epp has recently
shown that, among our fifty-four oldest codices of NT writings, 54 per cent
are gospels, and only 22 per cent are Pauline, so the use of a single codex
for Paul’s letters seems to have been less influential than Gamble’s theory
envisages.20
The circulation of Paul’s letters in a single codex and the four-gospel
codex and a variety of other factors I shall mention below are likely to
have helped to sustain early Christian addiction to the codex. But none of
the ‘big bang’ hypotheses provides an adequate explanation for the initial
adoption of the codex as the preferred format for early Christian writings.

8.2 the origin of the codex : three stages

How is early Christian ‘addiction’ to the codex to be explained? Recent
discoveries underline even more firmly the extent of the ‘addiction’. Twelve
further early fragmentary papyri of New Testament writings were published
between 1997 and 1999.21 All are in the codex format, not the roll which
was still being used almost universally for literary texts.
I shall work backwards, so to speak, from the point at which scribes copy-
ing non-Christian literary writings began to prefer the codex format to the
roll (my stage three). I shall then discuss the reasons why Christian scribes
became obsessed with the new format at a much earlier period (stage two).
18 Gamble, Books and Readers, p. 58. 19 See Chapter 3 above.
20 E. J. Epp, ‘The Codex and Literacy in Early Christianity and at Oxyrhynchus: Issues Raised by
Harry Y. Gamble’s Books and Readers in the Early Church’, Critical Review of Books in Religion 10
(1997) 15–37, esp. 35.
21 Oxyrhynchus Papyri, Vols. liv, lv, lvi. P. Oxy. 4403 and 4404 = ∏104–5, fragments of Matthew;
P. Oxy. 4445–8 = ∏106–8, fragments of John; P. Oxy. 4494–4500 = ∏110–15, fragments of
Matthew, Luke, Acts, Romans, Hebrews, and Revelation.
170 The gospels and papyri codices
Finally, I shall turn to the key question (stage one): how are we to explain
the initial and precocious use of the codex by Christian scribes? I shall argue
that the answer to this key question forces us to reconsider several common
assumptions concerning the transmission of early Christian traditions.
Before we proceed, a comment on the nature of the evidence and its
interpretation is necessary. There are gaps in our knowledge at key points,
and the limited evidence we have is often difficult to interpret. The evidence
takes three forms. References to rolls and codices in writers such as Martial
and Quintilian are often difficult to interpret. Iconographic depictions in
portraits, on vases, and on reliefs on sarcophagi are invaluable, but they
are often stylized and may have been influenced more strongly by icono-
graphical conventions than by ‘on the ground’ practice. Extant examples of
parchment and papyrus codices, and their predecessors, are difficult to date
with precision, for palaeography is an art and not a science. So disciplined
historical imagination and considerable caution are called for.
To complicate matters still further, there is varied use of terminology,
both in our primary sources and in the secondary literature. For example,
the Latin word codex is used both for a simple parchment or papyrus
notebook intended primarily for personal use and also for a more substantial
‘book’ intended to be read by others. Although some modern writers use
‘codex’ for both, I shall refer to the former as ‘notebook’ and the latter as
‘codex’.
Now back to my stage three. Up to about ad 300 very nearly all literary
texts were ‘published’ or copied in the roll format. But, by the fourth
century, only about 25 per cent still used the roll, and, by the fifth century,
the percentage had dropped still lower, to around 10 per cent.22 During
these centuries almost all Christian copies of Biblical manuscripts were in
the codex format; so too were most non-Biblical Christian manuscripts.
On the other hand, Jewish scribes continued to prefer the roll.
Although early Christian use of the codex has attracted lively discus-
sion, less attention has been given to the reasons why non-Christian scribes
eventually adopted the codex format. That important transition can be pin-
pointed to the decades around ad 300. Fresh light on that major change
has recently been shed by William V. Harris, well known for his work on
levels of literacy in antiquity.23

22 For the details, see Roberts and Skeat, Birth, pp. 36–7. More recent discoveries, or alternative datings
for some of the papyri, do not alter the general picture.
23 W. V. Harris, ‘Why did the Codex Supplant the Book-Roll?’, in J. Monfasani and R. G. Musto, eds,
Renaissance Society and Culture, FS E. F. Price Jr (New York: Italica Press, 1991), pp. 71–85.
 Why were early Christians addicted to the codex? 171
It is no coincidence, he argues, that the transition from roll to codex for
literary writings took place c. ad 300, ‘in a period when the numbers and
the respectability of the Christians were strongly increasing’. By the end of
the third century ‘no simple contrast could be drawn between the culture
of the Christians and the pagans. It was now possible to be both a Christian
and an ardent lover of the classics without any great strain.’24
If Harris is correct, it was scribes’ preference for the codex for Christian
writings which provided a major impetus for the use of this format for
‘secular’ literary writings c. ad 300. In line with the views of earlier writers,
Harris accepts that cost and ease of reference were vital advantages of the
codex form, as was its increased capacity; he lays special emphasis on ease
of reference as a key factor.
Only by about ad 300 is it plausible to postulate that Christian scribes’
addiction to the codex may have influenced non-Christian scribes. That is
inconceivable in the first and second centuries ad, when Christians were
scattered minority groups, somewhat isolated from the literary intelligentsia
of the day.
So much for stage three. We turn now to what I am calling stage two. In
The Birth of the Codex (1983), C. H. Roberts and T. C. Skeat played down
pragmatic reasons for acceptance of the codex format in order to leave room
for their own ‘big bang’ theories. Along with William Harris, Joseph van
Haelst,25 and E. J. Epp,26 I think that they were unwise to do so.
I am convinced that, once Christian scribes began to use the codex, their
precocious devotion to the new format was sustained by a variety of prag-
matic factors, some of which will have been more influential than others
at particular times and in particular locations.27 In Chapter 3, pp. 84–5,
first published in 1997, I discussed briefly the various pragmatic expla-
nations which have been offered and then set out my own. I noted that
early Christian codices, whether Roman or Christian, were quite small in
size and therefore much more portable than rolls. Christian scribes prepar-
ing writings to be carried by missionaries, messengers, and travellers over
long distances would have readily appreciated the advantages of the codex.
Their general counter-cultural stance would have made them more willing
24 Ibid., pp. 74 and 84.
25 J. van Haelst, ‘Les Origines du codex’, in Alain Blanchard, ed., Les Débuts du codex, Bibliologia 9
(Brepols: Turnhout, 1989), pp. 12–35. Van Haelst’s article is an extended critical appraisal of Roberts
and Skeat, Birth, with fresh evidence and discussion at many points.
26 Epp, ‘The Codex and Literacy’, pp. 15–37. On p. 24 he notes that, in 1996, he and I both used the
term ‘big bang’ theory within three months of one another, but quite independently.
27 I owe the phrase ‘precocious devotion’ to Harris, ‘Why did the Codex Supplant the Book-Roll?’,
p. 75.
172 The gospels and papyri codices
than their non-Christian counterparts to break with the almost unanimous
preference for the roll and experiment with the unfashionable codex. I also
noted that copying and using the Old Testament Scriptures and their foun-
dation writings in a new format was one of the ways Christians expressed
their sense of ‘newness’. Once the new format began to be adopted, its use-
fulness for collections of writings such as the four gospels and the Pauline
corpus would have enhanced its value.
E. J. Epp has recently added a further important consideration. The likely
content of the codices carried by early Christian missionaries and teachers
(whether Mark, OT testimonia, the Pauline corpus, the four-gospel codex)
is less important than the mere presence and use of the codex ‘in the highly
charged setting of evangelism and edification in pristine Christianity –
especially when a respected visitor is present with this new mark [i.e. a
codex] of his/her calling’.28
All these factors were instrumental in encouraging Christian scribes to
use the new codex format. But they all relate to what I now call stage two:
the rapid and almost universal acceptance of the new format by Christian
communities. How are we to explain stage one: the initial use of the codex
format by Christian scribes at a time when the papyrus or parchment roll
was the norm in society at large?
The roll held sway for a very long time. The oldest papyrus roll yet
discovered was found in an Egyptian tomb which dates from about 3000 bc.
To the archaeologists’ disappointment, it was blank, as it had been left ready
for the dead man’s use.29 The longest roll, written about 1150 bc, is about
43 cm (17 in) high and 40.5 m (133 ft) long. Leather rolls could also be very
large: the Temple Scroll, the largest of the Dead Sea Scrolls, is 8.2 m long
(27ft).30 A typical roll 18 cm (7 in) high and 6 m (20 ft) in length could
hold a single gospel or Acts, and could be held comfortably between thumb
and forefinger.31 The demise of the roll was slow: only in about ad 300 did
the codex reach parity with the roll for book production,32 and the roll
continued to be used for formal documents right up to medieval times and
beyond.

28 Epp, ‘The Codex and Literacy’, p. 21.

29 Alan Millard, Reading and Writing in the Time of Jesus (Sheffield: Sheffield Academic Press, 2000),
p. 24.
30 Ibid., pp. 61–2.
31 Roberts and Skeat, Birth, p. 47. For a good summary of the methods used to make papyrus and
parchment rolls, see Gamble, Books and Readers, pp. 43–8.
32 For details of the approximate percentages from the first to the fifth century, see Roberts and Skeat,
Birth, p. 36.
 Why were early Christians addicted to the codex? 173
When and why was the codex first used? Since ‘nothing comes from
nothing’, the codex must have had predecessors. So, before we consider the
origin of the codex, we must discuss its predecessors.

8.3 predecessors of the codex

The ubiquitous roll was obviously one of the predecessors of the parchment
or papyrus codex, but there were others.33
(1) Wooden (or occasionally ivory) tablets had a rim around the edge, to
hold wax in place.34 They were incised with a stylus; they could be readily
reused by smoothing over the wax. The wax tablets were used for notes
and drafts of various kinds, including school exercises. Several wax tablets
could be held together by a cord which passed through holes in the tablets.
Codex or caudex were the Latin words for a set of tablets held together in
this way; d”ltov or p©nax the usual equivalent terms in Greek.35
Wooden wax-covered tablets are depicted in a number of portraits and
sculptures, most notably in several wall paintings at Pompeii. Like the roll,
they have a long history. From the second half of the second millennium bc
wax-covered wooden tablets were widely used in the Near East. They have
been discovered in places as far apart as Egypt and the north of England.
A large number were found at Herculaneum and at Pompeii.36 Up to ten
tablets were sometimes tied together, but four would usually have been
a handful. Writing in ad 85, the Roman poet Martial referred to a ‘five-
leaved wax tablet’ (quinquiplici cera), a ‘three leaved tablet’, ivory tablets,
and even waxed parchment which could be erased.37 Wax tablets could be
readily reused by smoothing over the wax (perhaps slightly warmed) with
a spatula. Occasionally both sides were inscribed.
A particularly interesting example of a double-sided wax tablet was un-
earthed in the 2001 archaeological season at Vindolanda, near Hadrian’s
33 An earlier version of parts of sections 8.3 and 8.4 was published in G. N. Stanton, ‘The Early
Reception of Matthew’s Gospel’, in D. E. Aune, ed., The Gospel of Matthew in Current Study (Grand
Rapids: Eerdmans, 2001), pp. 53–60.
34 For a wide-ranging, well-illustrated discussion of inorganic and organic materials used in antiquity
for writing, see Horst Blanck, Das Buch in der Antike (Munich: C. H. Beck, 1992), pp. 40–63.
35 On the difference between codex (referring to wooden tablets) and p©nax, see M. Haran, ‘Codex,
Pinax and Writing Slat’, Scripta Classica Israelica 15 (1996) 212–22. He claims that, whereas the former
was Roman and not mentioned in the Talmud, the p©nax had a Greek-Hellenistic background and a
concertina-like form, and was occasionally used for other purposes such as serving food. The p©nax
is referred to in several Talmudic passages, but it was never made of papyrus.
36 For details and full references see Millard, Reading and Writing, pp. 26–9. See also A. K. Bowman
and J. D. Thomas, Vindolanda: The Latin Writing-Tablets, Britannia Monograph Series 4 (London:
Society for the Promotion of Roman Studies, 1983), pp. 33–5.
37 Martial, Epigrams xiv.4–7.
174 The gospels and papyri codices
Wall in northern England. This tablet, which dates from c. ad 100, was in-
scribed not only with a stylus pen on the wax (which has long disappeared),
but also with pen and ink on the rim of the tablet, to assist filing.38
Like the roll, wax tablets remained in use long after the codex came into
general use. For example, they are listed as essential equipment for a monk
in the fifth-century Rule of St Benedict. A wax tablet is described in a riddle
written by Aldhelm, bishop of Sherborne in the latter part of the seventh
century:
Of honey-laden bees I first was born, but in the forest grew my outer coat; my
tough backs came from shoes (the leather thongs used to tie the tablets into a
notebook). An iron point in artful windings cuts a fair design, and leaves long
twisted furrows like a plough.39
From the same period we have an interesting example of the use of wax
tablets in the planning of a book. Adamnan, Abbot of Iona 679-704, wrote
an account of the Holy Places in Palestine. In his preface he tells us that
wax tablets were used to note down his pilgrim informer’s experiences and
to make sketch plans of the places he had visited.40
(2) Until recently it was generally assumed that wooden wax-covered
tablets were the standard writing-material for letters and documents in
those parts of the Roman Empire where papyrus was not readily available.
However, the publication in 1983 and 1994 of the very large number of thin
leaf tablets from Vindolanda (Northumberland, England) which date from
c. ad 100 necessitates a reappraisal of this view. The Vindolanda discoveries
have not yet received the attention from New Testament scholars which
they deserve. They came too late for more than passing mention in Joseph
van Haelst’s major study41 and in Harry Gamble’s survey.42
The Vindolanda tablets are ‘thin slivers of smooth wood which are writ-
ten with pen and ink’, and may conveniently be referred to as ‘leaf tablets’.43
Most were used for official documents, as well as for letters and drafts of
letters. However, at least one of the Vindolanda tablets contains a literary
text, a line of Virgil; three others may be literary or semi-literary.44 In the

38 This information comes from a display cabinet in the museum at Vindolanda. At the time of
writing, the Report of the 2001 excavations had not yet been published, nor had the promised full
publication of all the wax tablets discovered at Vindolanda taken place. For up-to-date information,
see the Vindolanda web site: www.vindolanda.com.
39 Quoted by Janet Backhouse, The Lindisfarne Gospels (London: Phaidon, 1981), p. 31.
40 Ibid., pp. 31–2. 41 Van Haelst, ‘Les Origines du codex’ (above, n. 25), pp. 15–16.
42 Gamble, Books and Readers, p. 268 n. 35. 43 Bowman and Thomas, Vindolanda, p. 32.
44 A. K. Bowman, Life and Letters on the Roman Frontier: Vindolanda and its People (London: British
Museum, 1994), p. 18.
 Why were early Christians addicted to the codex? 175
light of their publication, it is probable that some of the many literary refe-
rences in first- and second-century writings to notebooks (pugillaria) may
be to leaf tablets rather than to wooden wax-covered tablets, also referred
to as stylus tablets.
In 1983, the editors of the Vindolanda tablets accepted that for two
reasons they could not be described as a primitive codex: some of them
were linked together in ‘concertina’ format, and with only a couple of
partial exceptions not both sides of the leaf had been written on. However,
they also noted that the existence of this wooden notebook in this format
at a period which was clearly an important one for the emergence of the
codex may be of some significance for that development.45
In 1994 the tablets were re-edited, together with the considerable finds
from the 1985–9 excavations. The editors noted that there was no standard
format for documents, though a diptych format, with the address written
on the back of the right-hand half of the diptych, is the norm for letters.46
Somewhat surprisingly, the editors did not comment further on the rela-
tionship of the leaf tablets to the origins of the codex. But there can now be
little doubt that, with the very thin Vindolanda leaf tablets from c. ad 100,
inscribed with pen and ink, we have extant examples of notebooks with a
strong a claim to be one of the forerunners of the codex.47
By 2001 a total of 1,400 or so leaf tablets had been discovered at Vin-
dolanda, but only about 300 wax-covered stylus tablets. However, some
230 stylus pens have been found, so perhaps the numbers are misleading.48
The 2001 excavations unearthed 35 leaf tablets and 17 stylus tablets.
A very small number of similar leaf tablets have been found in other parts
of Britain, in Dacia, and in Vindonissa (Switzerland). One found among
the Bar Kochba letters is especially important. It contains a letter from
‘Shimo’on Bar Kosiba, the Prince of Israel, to Yehonatan and Masabalda’
concerning three matters. The thin wooden slat, written in ad 134 or 135,
has been incised down the middle, so that it could be folded. There are nine
lines of writing on the right side, and eight on the left. The editor, Y. Yadin,
notes that the practice of writing on wood was widespread throughout the

45 Bowman and Thomas, Vindolanda, pp. 40–4.

46 A. K. Bowman and J. D. Thomas, The Vindolanda Writing-Tablets: Tabulae Vindolandenses ii
(London: British Museum, 1994), pp. 40–6.
47 Bowman, Life and Letters, p. 10 notes that, with writing-tablets still coming out of the ground in
the 1990s, conclusions can only be tentative and provisional.
48 In an oral presentation to an SNTS seminar in Durham in August 2002 Professor J. D.
Thomas mentioned that, whereas wax tablets made excellent kindling, leaf tablets would not burn
readily.
176 The gospels and papyri codices
Orient, and is often mentioned even in rabbinical literature.49 The signature
does not correspond with the handwriting in the letter itself: as in one of
the Vindolanda letters (the so-called Claudia Severa letter), and as Paul
himself mentions at Gal. 6.11, the writer has dictated the letter and only
then ‘signed’ it himself.
Unlike the wax-covered wooden tablets, the leaf tablet or slat could not be
readily reused. They were too thin and too flimsy to allow ink to be scraped
off. However, they were light and could be linked together, sometimes in
concertina fashion. It was possible to write on both sides – and sometimes
the reverse side was used, especially for the addressee of a letter. It is worth
noting that there was a degree of overlap between the wax-covered tablet and
the leaf tablet, and that the same Latin word, pugillaria, seems to have been
used for both.50 The rim of some of the Vindolanda wax tablets is marked
with ink, either in order to give a summary of the contents, or to assist the
process of filing. The wax was often incised so firmly that marks were left
on the wood underneath. Although the wax has long since disappeared,
some of the writing on these tablets can still be read; sometimes the writing
comes from more than one document. Occasionally ink has been used on
Vindolanda stylus or wax tablets.51
Although there are some lines of continuity from both the wax-covered
stylus tablets and the leaf tablets or slats to the codex, there are also obvious
differences. As with the codex, in both cases the materials used for writing
were in a small ‘page’ format; occasionally the reverse side was also used.
Although individual ‘pages’ were often linked together to form a ‘booklet’,
the end result was clumsy and unsuitable for more than notes, drafts, and
sketches.
(3) The third and much more direct predecessors of the codex were
parchment or papyrus notebooks (membranae, membr†nai).52 Although
the earliest extant examples are from the end of the second or the third
century ad, there is literary evidence which confirms that they were used
much earlier.53
49 Y. Yadin, ‘Expedition D’, Israel Exploration Journal 11 (1961) 41–2. See also Haran, ‘Codex, Pinax
and Writing Slat’, 212–22.
50 Bowman and Thomas, Vindolanda, p. 43.
51 See Robin Birley, Roman Records from Vindolanda on Hadrian’s Wall, Roman Army Museum
Publications for the Vindolanda Trust, (Carvoran: Greenhead, 3rd edn, 1999), pp. 21–2.
52 Cf . J. van Haelst: ‘Le carnet de parchemin est une étape intermédiaire indispensable entre la tablette
de cire et le codex. Ce sont ses feuillets qui, multipliés selon les besoins, pourront éventuellement
contenir une œuvre littéraire de quelque étendue.’ ‘Les Origines du codex’, p. 20.
53 See Roberts and Skeat, Birth, pp. 15–23 and van Haelst, ‘Les Origines du codex’, p. 18 for details and
bibliography. Part of a probable parchment notebook has been found at Dura-Europos and dated
to the first century bc or ad. See Bowman and Thomas, Vindolanda, p. 43.
 Why were early Christians addicted to the codex? 177
Quintilian’s comments from about ad 90 are particularly important:
It is best to write on wax owing to the facility which it offers for erasure, though
weak sight may make it desirable to employ parchment pages (membranae) by
preference . . . But whichever we employ, we must leave blank pages that we may
be free to make additions when we will. (Inst. Or. x.3.31-2)54
Wax tablets and parchment pages are quite clearly differentiated. The refe-
rences to pages on the left side being left blank and to the better visibility
of writing with ink on parchment confirms that Quintilian is well aware of
the considerable advantages of the notebook. The use of parchment does
not seem to be an innovation, yet one senses from this passage that the wax
tablet is the norm in Quintilian’s day.
There is only limited evidence for the existence of parchment or papyrus
notebooks earlier in the first century. The words attributed to Paul in 2
Tim. 4:13 provide important support for their existence a decade or two
before Quintilian wrote. ‘When you come’, ‘Paul’ writes to Timothy, ‘bring
the (traveller’s) cloak I left with Carpus at Troas, and the books in roll
format (t‡ bibl©a), particularly my parchment notebooks (m†lista t‡v
membr†nav)’.55 This sentence has elicited plenty of discussion.56
Are two groups of writings referred to, as in my translation, or only one?
T. C. Skeat gained only minimal support for his view that one group of
writings is referred to here, parchment notebooks: he took m†lista as ‘to
be precise, namely, I mean’.57 However, most scholars accept that t‡ bibl©a
is a general reference to writings, almost certainly in roll format, and that
t‡v membr†nav refers to writings in a different format, about which the
writer is especially anxious.
What does t‡v membr†nav refer to? Membr†na is a Latin loan-word
transliterated into Greek, for there was no Greek word available.58 We

54 ‘Scribi optime ceris, in quibus facillima est ratio delendi, nisi forte visus infirmior membranarum
potius usum exiget . . . Reliquendae autem in utrolibet genere contra erunt vacuae tabellae, in quibus
libera adiciendi sit excursio.’
55 See Michael McCormick, ‘Typology, Codicology and Papyrology’, Scriptorium 35 (1981) 331–4.
56 See especially the discussion and references in I. H. Marshall, The Pastoral Epistles, ICC (Edinburgh:
T. & T. Clark, 1999), pp. 818–21.
57 T. C. Skeat, ‘Especially the Parchments: A Note on 2 Timothy iv.13’, JTS 30 (1979) 173–7. See also
Roberts and Skeat, Birth, p. 22. This possible interpretation of m†lista is not even mentioned in
BDAG. However, E. R. Richards, The Secretary in the Letters of Paul (Tübingen: Mohr, 1991), p. 164
n. 168 allows that Skeat may be correct, referring to Gal. 6.10; I Tim. 4.10; also Phil. 4.22; Philem.
16; I Tim. 5.8, 17; Titus 1.10.
58 See further van Haelst, ‘Les Origines du codex’, p. 17. He notes that the term kädix, a transliteration
of the Latin ‘codex’, is found in the Byzantine era, but it does not refer to a book containing literary
works, whether ecclesiastical or secular; it refers to a register of taxes, or a collection of documents,
inventories, or archives.
178 The gospels and papyri codices
have just noted that Quintilian used membrana to refer to a parchment
notebook, and this is clearly the meaning here.59
If only we knew what those parchment notebooks contained! The con-
text gives us some clues, but not a clear answer. ‘Paul’ has just reminded
Timothy of the importance of the Scriptures and has provided a list of the
uses to which the Scriptures should be put (2 Tim. 3.14-17). So it is proba-
ble that t‡ bibl©a includes rolls of OT Scriptures. How many rolls could
Timothy have been expected to carry, along with that traveller’s cloak, and
his own personal possessions? Surely not many. So, in order to supplement
a small number of rolls, parchment notebooks may well have contained
excerpts from Scripture, perhaps also some ‘trustworthy sayings’ (cf. II Tim.
2.11) and some drafts of letters. We shall return to this important point
below.
It is not surprising that examples of parchment or papyrus notebooks
from the first century have not survived. These notebooks were used for
letters, and for ephemeral notes and documents of various kinds, but not
for writings which might be treasured by a later generation, for which the
carefully constructed roll was the norm. Although Quintilian and 2 Tim.
4.13 refer to parchment notebooks, it is probable that papyrus notebooks
existed at the same time, particularly in Egypt and the East, where papyrus
was readily available; we should not suppose that the one evolved out of
the other.60 The two types of notebook found at Vindolanda confirm that
local availability dictated the material used. On the northern frontier of the
Roman Empire, papyrus would not have been available, and parchment
would have been difficult to prepare. The huge numbers of wax stylus
tablets and ink leaf tablets (more than 1,700 in all by 2002) found in the
sodden earth at Vindolanda call in question current assumptions about
very limited levels of literacy in antiquity.
The evidence for the widespread use of the different types of notebooks
confirms that, for many people in the Graeco-Roman world, notebooks
were part of everyday life. Although we are still unable to reconstruct
with confidence how, why, and when the more substantial papyrus and
parchment codex evolved out of the various kinds of notebooks mentioned
above, a gradual evolution is more probable than any one ‘trigger’ factor.61
59 Pliny informs us that his uncle used to work with a slave standing by him holding a book and
notebooks (libro et pugillaribus). Ep.iii.5.15.
60 See E. G. Turner, ‘The Priority of Parchment or Papyrus?’, in his The Typology of the Early Codex
(Philadelphia: University of Pennsylvania Press, 1977), pp. 35–54. Turner challenges the older view
that parchment codices preceded papyrus codices.
61 Roberts and Skeat, Birth, p. 10 correctly note that the transition from papyrus to parchment was of
an entirely different character from, and quite unconnected with, the transition from roll to codex.
 Why were early Christians addicted to the codex? 179
8.4 the f irst cod ices for ‘literary’ writings
By no later than the end of the first century ad the codex was being used by
a small number of non-Christian writers for more substantial writings than
notes, documents, drafts, and letters. As we shall see in a moment, some
of these writings were literary: it is a mistake to assume from the origins
of the codex that this format was reserved solely for utilitarian writings or
handbooks.62
At present there are two rivals for the accolade of the earliest extant
codex: P. Oxy. 30, a parchment codex in Latin from about ad 100 which is
usually known as De Bellis Macedonicis,63 and ∏52, the well-known papyrus
fragment of John 18, which is generally dated to about ad 125. Although
P. Oxy. 30 is fragmentary, it is clearly a historical writing. It is no mere set
of ephemeral notes – and neither is John’s Gospel!
Was the codex format first used by Christians, and only then adopted by
non-Christian scribes? The earliest extant examples just noted do not settle
the question. Their palaeographical dating is not certain, and in any case
there is literary evidence for earlier non-Christian use of the codex which
must be weighed carefully.
Writing in ad 85, the Roman poet Martial refers to the availability of
parchment codices for travellers: ‘pocket editions’ of Homer, Virgil, Cicero,
Livy, and Ovid which are referred with the words in membranis (in parch-
ment page format) or in pugillaribus membraneis (hand-held parchment
pages; Epigrams i.2 and xiv.184–92). In Martial’s terse two-line mottoes
about these ‘pocket editions’ there is no hint that they are mere novelties,
or a recent innovation. In their influential The Birth of the Codex, C. H.
Roberts and T. C. Skeat claimed that Martial’s ‘experiment was still-born’.64
In an equally memorable and in my judgement valid comment on that view,
Joseph van Haelst is dismissive: ‘C’est excessif.’65
We do have a sprinkling of codices of non-Christian writings from the
second century. Although they make up only 2 per cent of the total (the
remainder are on rolls), they should not be set aside as an insignificant
aftermath of a still-born experiment. One of the second-century codices
is a parchment in Latin (P. Oxy. 30, noted above). Three are parchment

They allow the possibility that the papyrus codex and the parchment codex may have developed in
parallel (p. 29).
62 Roberts and Skeat, ibid., p. 5 n. 1 note that it is quite wrong to describe the papyrus codex as a
‘bastard form’.
63 Turner, Typology, pp. 38, 93 dates this to the early second century; J. Mallon (cited by Turner, p. 128),
to the end of the first century.
64 Roberts and Skeat, Birth, p. 29. 65 Van Haelst, ‘Les Origines du codex’, p. 21.
180 The gospels and papyri codices
codices in Greek; fourteen are papyrus codices in Greek. Many are literary
writings.66
P. Petaus 30 (first published in 1969) is particularly important.67 This
letter in Greek, which can be dated confidently to the second century, but
not more precisely, refers to eight parchment codices which were purchased,
and six more which were not.
Julius Placidus to his father, Herclanus, greetings.
Deios has been with us and has shown us six parchment codices (t‡v membr†nav
™x). We didn’t select any of them, but we did collate eight others, for which I paid
on account a hundred drachmas . . . I wish you good health.
P. Petaus 30 ‘implies a touring book-seller offering literary membranae’.68
As with Martial, there is no suggestion that these readily portable codices
were an innovation. The social setting reflected in the letter is important:
these codices were part of a mobile bookshop. Martial, it will be recalled,
refers to pocket editions of literary codices (membranae) which travellers
would find useful. The earliest non-Christian codices seem to have been
used for a variety of writings, including literary writings in the ‘classical
canon’ of the day. As far as we can judge, they first became popular with
travellers.69
Taken together, the extant and the literary evidence suggest that codices
for substantial writings were not unknown in non-Christian circles in the
latter half of the first century, i.e. at the time of the composition of the New
Testament writings. At this time it is most unlikely that invention of the
codex format by Christian scribes would have been imitated and developed
by non-Christian scribes, albeit in a limited way.70 The reverse is possible:
for their own reasons, Christian scribes may have imitated an innovation
by a small minority of their secular counterparts. Or, Christian and non-
Christian scribes may have begun to experiment with more substantial

66 See the slightly different lists with full details in Roberts and Skeat, Birth, p. 71 and in van Haelst,
‘Les Origines du codex’, pp. 23–5.
67 For the critical edition, see U. and D. Hagedorn and L. C. and H. C. Youtie, eds., P. Petaus: Das
Archiv des Petaus, Pap. Colon., Vol. iv (Cologne and Opladen: Westdeutscher Verlag, 1969).
68 E. G. Turner, Greek Papyri, 2nd edn (Oxford: Clarendon, 1980), p. 204.
69 On the basis of his analysis of the second-century non-Christian codices, Michael McCormick
suggests that the novel format appealed to some ancient doctors and teachers. As they were often on
the move, they would have found the codex format handy. ‘Typology, Codicology and Papyrology’,
pp. 331–4.
70 Several more recent writers have been reluctant to accept the view of Roberts and Skeat, Birth, that
Christians invented the codex. See, for example, Robin Lane Fox, ‘Literacy and Power in Early
Christianity’, in A. K. Bowman and Greg Woolf, eds., Literacy and Power in the Ancient World
(Cambridge: Cambridge University Press, 1994), p. 140; Epp, ‘The Codex and Literacy’ (above,
n. 20), pp. 15–16, who states that he is now more hesitant than in some of his earlier publications.
 Why were early Christians addicted to the codex? 181
notebooks, i.e. codices, at about the same time, but quite independently;
only in Christian circles did an addiction to the new format take hold
and become the norm rather than the exception. I think that the latter
possibility is more likely.
At present the consensus view is that the autographs and the very earliest
copies of Christian writings were on rolls, and that, at the end of the first
century or thereabouts, Christian scribes switched to the codex under the
influence of a hypothetical ‘big bang’. We have seen that all the ‘big bang’
theories on offer fail to convince. Is there a more plausible reason why
Christian scribes first turned to the codex? As noted above, several factors
seem to have sustained and strengthened addiction to the codex (my stage
two), but how did it take hold in the first place (my stage one)? And is it
possible that the codex was used for the very first Christian writing?

8.5 early christian not ebooks

We have seen that, in the first century, several kinds of notebooks were the
predecessors of the codex. They were used for all manner of ‘non-literary’
writings all over the Roman Empire, from the Near East to Vindolanda
in the far North-West. As we have seen, in the latter part of the first cen-
tury, the same word membranae was used in Latin and in Greek both for
a notebook (Quintilian; 2 Tim. 4.13) and for a ‘literary’ codex (Martial;
P. Petaus 30). With hindsight, the use of codices for more permanent writ-
ings than ephemeral notes was a very natural development. However, pref-
erence for the roll for literary writings was so deeply entrenched that a
change of format was anything but obvious to most in the first century.
In the post-Easter decades the followers of Jesus would almost certainly
have made use of more than one kind of notebook: in some settings wax
tablets, in others, wooden leaf tablets or slats, and, in others, papyrus or
parchment notebooks – as well as, of course, single sheets. Their regular
use for varied purposes may have encouraged Christians to make use of
more substantial notebooks, i.e. codices, for their more permanent writings.
And, since Christians formed minority, partly counter-cultural groups with
limited literary pretensions, the use of the codex rather than the roll did
not involve a major shift in mind-set. In short, a gradual evolution from
‘notebook’ to ‘codex’ is much more likely than a big bang development
ex nihilo. This explanation for initial Christian use of the codex (my stage
one) is more plausible than any of the alternatives which have been offered.
The theory is both simple and elegant! But where is the evidence?
182 The gospels and papyri codices
I must confess immediately that, with the exception of 2 Tim. 4.13, we
do not have explicit evidence for use of notebooks by first- or even early
second-century Christians. But surely Paul’s first letter to the Thessalonians
(c. ad 50) was not the very first time when writing was used by a Christian.
By then some form of notebook may well have been used for a couple of
decades for several different Christian purposes. I shall now refer to three
possibilities.

8.5.1 Scriptural excerpts and testimonies

Luke records that Jesus read from the scroll of the prophet Isaiah in his
home-town synagogue in Nazareth (Luke 4.16-20). Here NRSV and most
modern translations correctly use the term ‘scroll’ or ‘roll’, abandoning the
translation ‘book’ used in the 1611 AV/KJV, and even in the 1946 RSV. But
what happened in the immediate post-Easter period?
We know very little about the 30s and 40s of the first century, but we have
enough evidence to confirm that in those decades Christian missionaries
or teachers did not always have ready access to local synagogues in order to
consult rolls of the Scriptures. In any case, few of them could read. Making
their own copies was time-consuming and expensive. And carrying handfuls
of rolls of favourite Christian Scriptures such as the Psalms and Isaiah on
their often arduous journeys was not easy. So in all probability some kind of
notebook was used for Scriptural excerpts which were prominent in early
Christian preaching and teaching.
Early Christian use of written collections of Scriptural excerpts and tes-
timonia, i.e. proof-texts to support Christian claims concerning Jesus, is
now attracting renewed scholarly interest. In 1889 Edwin Hatch suggested
that the early Christians took over the Jewish practice of compiling ex-
tracts of Biblical passages, but his proposal attracted little attention.71 In
1916 and 1920 J. Rendel Harris proposed that there existed in the early
church a single Testimony Book, a compilation of Scriptural passages drawn
on by several early Christian writers as ‘testimonies’ to underpin their con-
victions.72 Although his work attracted plenty of attention, most scholars
considered that the theory was unduly speculative. However, it is worth
noting that Rendel Harris’s views were less rigid than many of his critics

71 E. Hatch, Essays in Biblical Greek (Oxford: Clarendon, 1889). I owe this reference to T. H. Lim,
Holy Scripture in the Qumran Communities and the Pauline Letters (Oxford: Clarendon, 1997),
pp. 151–2.
72 J. Rendel Harris, Testimonies (2 vols., Cambridge: Cambridge University Press, 1916–20).
 Why were early Christians addicted to the codex? 183
supposed: he allowed for editions, expansions, and changes in the content
of the Testimony Book.73
The turn of the tide towards a more favourable view of the hypothesis
came with the publication in 1956 of 4Q Testimonia (now known as 4Q175)
and then of J. A. Fitzmyer’s fine study, ‘“4Q Testimonia” and the New
Testament’ the following year.74 On a single leather sheet probably compiled
by an individual for his own reference, several OT passages are juxtaposed.
4Q175 is a small Jewish anthology of related Scriptural texts. George Brooke
has shown convincingly that they refer to the eschatological figures of the
prophet, royal Messiah, and priest.75
4Q176 (also known as 4Q Tanhumim) is also important for our present
purposes. A large number of fragments of excerpts from Isaiah 40–55 have
been found. The excerpts all relate to Yahweh’s promises of ‘comfort’ for
his afflicted people.76 So we now have clear evidence that shortly before
the first century ad excerpts of Scriptural passages were gathered together
thematically.
There is an often-overlooked extant early Christian equivalent. The
publication of 4Q175 has led to renewed interest in Papyrus Rylands Greek
460 (PRG 460), first published by C. H. Roberts in 1936, and recently
examined in detail by Alessandro Falcetta.77 Here we have a cluster of OT
LXX-related passages, juxtaposed without comment on two leaves in sim-
ple codex format. The thematic linking of passages is strikingly reminiscent
of 4Q175. Although this papyrus probably dates only from the early fourth
century, PRG 460 provides clear evidence of early Christian use of written
collections of related Scriptural texts. Falcetta has proposed that PRG 460
originally contained four sections of texts, each occupying one ‘page’ of the
two leaves written on both sides. ‘We may think that other leaves, similar
in form to this one, were stitched together in order to make a book of
testimonia. Each leaf may have dealt with a particular topic.’ In the light
of 4Q175, a single sheet which does deal with one topic, Falcetta’s theory
seems to me to be very plausible indeed.
73 See Martin C. Albl, “And Scripture Cannot be Broken”: The Form and Function of the Early Christian
Testimonia Collections (Leiden: Brill, 1999), p. 20.
74 J. A. Fitzmyer, TS 18 (1957) 513–37; reprinted in J. A. Fitzmyer, Essays on the Semitic Background of
the New Testament (London: Chapman, 1971), pp. 59–89.
75 G. Brooke, Exegesis at Qumran: 4QFlorilegium in its Jewish Context (Sheffield: JSOT, 1985).
76 See C. D. Stanley, Paul and the Language of Scripture (Cambridge: Cambridge University Press,
1992), p. 76–7, and, more fully, his ‘The Importance of 4Q Tanhumim (4Q176)’, RevQ 15/60 (1992)
569–82.
77 C. H. Roberts, ‘Two Biblical Papyri in the John Rylands Library, Manchester’, BJRL 20 (1936) 241–4;
A. Falcetta, ‘A Testimony Collection in Manchester: Papyrus Rylands Greek 460’, BJRL 83 (2001)
3–19.
184 The gospels and papyri codices
PRG 460 is most unlikely to have been the only early Christian written
collection of Scriptural texts. Martin C. Albl has mounted an impressive
cumulative argument for their existence in the early post-Easter period.
He carefully distinguishes between testimonia collections ‘which serve foren-
sically to “prove” a certain point’ and other collections of Scriptural passages
used in the early church for a variety of purposes.78 Albl provides plenty of
Graeco-Roman and Jewish evidence for the popularity of written excerpts
of valued writings,79 and discusses at length the evidence for testimonia
collections in the patristic era and in the NT writings themselves.
Two strands of Albl’s argument are especially important for the present
discussion. He sets out criteria which allow the detection of written collec-
tions of Scriptural passages as follows: ‘(1) quotations that deviate con-
siderably from known scriptural texts (LXX or LXX recensions, MT),
especially in a Christianizing direction; (2) composite quotations; (3) false
attributions; (4) use of the same series of texts in independent authors;
(5) editorial or interpretive comments indicative of a collection; (6) evident
lack of awareness of the biblical context of a quotation; (7) use of the same
exegetical comments in independent authors’.80
Timothy Lim has recently defended a similar view. He notes that, in the
citation of Isaiah in Luke 4.18-19, parts of Isa. 61.1 and 58.6 are quoted in
that order. ‘It is possible that Jesus read the verses of Isa. 61 and 58 not from
a biblical text of Isaiah, but from a collection of excerpts used in liturgy.’81
Lim’s main point stands even if one is less confident that Luke 4.18-19 goes
back to Jesus.
If written collections of Scriptural excerpts existed among followers of
Jesus even before Paul wrote I Thessalonians c. ad 50, they would have
been in one of the forms of notebook referred to above. Since testimonia
and Scriptural excerpts played such an important role in the very earliest
Christian communities, the familiar notebook or ‘page’ format used is
likely to have been retained and developed as a form of codex for more
substantial writings rather than the roll.82 So we have one ready explanation
for the initial early Christian use of the codex format.
78 Albl, ‘And Scripture Cannot be Broken’, p. 6.
79 See also the examples noted by Stanley, Paul and the Language of Scripture, pp. 74–6.
80 Albl, ‘And Scripture cannot be Broken’, p. 66.
81 Lim, Holy Scripture in the Qumran Communities, p. 55.
82 So too Albl, ‘And Scripture Cannot be Broken’, pp. 98–9. My own earlier comments along very
similar lines published in 1997 are included above, p. 84. Albl and I seem to have reached the same
conclusion independently. H. Y. Gamble also allows that the codex may first have been used by
Christians for Scriptural testimonia, but claims that it was the collection of Paul’s letters early in the
second century (and probably earlier) which set the standard for use of the codex; Books and Readers,
p. 65.
 Why were early Christians addicted to the codex? 185
Interest in early Christian use of Scripture has intensified in recent
decades in the wake of the publication of the Dead Sea Scrolls. But few
scholars have paused to ask about the means by which the very earliest
post-Easter followers of Jesus gained and retained their knowledge of the
Scriptures, which they reinterpreted so creatively in the light of their new
convictions.83 Access to rolls of Scripture in synagogues was unlikely to
have been granted readily to many. No doubt favourite passages of Scrip-
ture were eagerly memorized and transmitted orally, but a ‘pan-oral’ theory
takes no account of the prevalence of written excerpts of ‘classical’ writings
in the Graeco-Roman, Jewish, and early Christian worlds.

8.5.2 Drafts and copies of letters

In one of his letters to his friend Paetus, Cicero writes as follows (46 bc): ‘I
have just taken my place at table at three o’clock, and am scribbling a copy
of this letter to you in my notebook’ (‘ad te harum exemplum in codicillis
exeravi’) (Ad Fam. 9.26.1).84 The content of the letter is not especially
noteworthy, yet Cicero keeps a copy in a notebook as a matter of course.
This was by no means unusual.
If copies of Paul’s letters were kept in notebook format, as were Cicero’s,
then there would have been a powerful incentive to use the developed form
of the notebook, the codex, in the first collection of the letters, which was
probably made by a co-worker shortly after Paul’s death.85 For a collection
of the letters made in, say, ad 85, a codex would have been more suitable
than a roll, for a standard-sized roll would not have held all the letters.86
Drafts of more formal ‘literary’ letters were usually transferred from
notebooks to parchment or papyrus rolls. But this may not have happened
with Paul’s letters – perhaps not even with his more substantial letters. Paul
and his colleagues were engaged in frenetic missionary work and may have
had other priorities. I think that we have to consider the possibility that at
least some of Paul’s letters originally circulated in a notebook format rather

83 Stanley, Paul and the Language of Scripture, pp. 74–5, is an exception. He suggests that, as Paul ‘came
across passages that promised to be useful later on, presumably he copied them down onto his handy
wax tablet, or perhaps even directly onto a loose sheet of parchment’.
84 I owe this reference to Richards, The Secretary (above, n. 57), p. 165 n. 169. His thorough examination
of the primary evidence for the conventions of letter-writing shows that ‘copies of letters were a
desirable thing in the ancient world’ (p. 6). See also David Trobisch, Paul’s Letter Collection: Tracing
the Origins (Minneapolis: Fortress, 1994), pp. 56–7.
85 Trobsich, Letter Collection, argues that Paul himself collected together Romans, 1–2 Corinthians, and
Galatians, edited them, and sent them as a unit to friends in Ephesus. Although many interesting
points are made, the theory fails to convince.
86 So too Gamble, Books and Readers, pp. 62–3.
186 The gospels and papyri codices
than on rolls. Given the widespread use of notebooks referred to above, it is
very probable that Paul and his co-workers and secretaries (and other early
Christian letter-writers) used notebooks for preliminary notes and drafts
and for copies of letters, as well as for collections of Scriptural texts.87
Although there is now greater willingness to allow that Paul’s use of a
secretary is relevant to authorship questions,88 use of notebooks is rarely
considered. I think that it is relevant to discussion of several of the complex
problems thrown up by Paul’s letters. For example, how are we to account
for the juxtaposition in Ephesians of both Pauline and non-Pauline strands?
John Muddiman has recently reopened this question by proposing that an
authentic Pauline letter (to the Laodiceans) was subsequently edited and
expanded, with the aim of adjusting the Pauline tradition to the situation
prevailing at its time and place of composition.89 Perhaps this attractive
theory is made even more plausible if we bear in mind the possibility that
the original, authentic letter only ever existed in draft form in a notebook –
a draft which was expanded at a later point by one of Paul’s co-workers or
close followers.

8.5.3 Jesus traditions?

Were Jesus traditions also transmitted, at least in part, in some form of
notebook? Even to pose the question is to court derision from scholars
wedded to the view that, right up to the time Mark wrote his gospel,
Jesus traditions were always transmitted orally, and only orally.90 However,
occasionally voices to the contrary have been raised; this is certainly a
question which should be reopened.
In 1950 the respected Jewish specialist on the rabbinic writings, S. Lieber-
man, claimed that Jesus’ disciples wrote down the sayings of Jesus:
Now the Jewish disciples of Jesus, in accordance with the general rabbinic practice,
wrote the sayings of their master, pronounced not in the form of a book to be
published, but as notes in their pinakes, codices, or in their note-books (or in

87 So also Richards, The Secretary, pp. 161–5, 191, and K. P. Donfried, ‘Paul as SKHNOPOIOS and the
Use of the Codex’, in his Paul, Thessalonica and Early Christianity (London: T. & T. Clark, 2002),
pp. 293–304.
88 See especially Richards, The Secretary, passim.
89 John Muddiman, The Epistle to the Ephesians, Black’s NT Commentaries (London and New York:
Continuum, 2001). A translation of the tentatively reconstructed authentic letter is printed as
Appendix B, pp. 302–5.
90 For a vigorous rejection of Werner Kelber’s claim that Mark’s first written gospel was intended
to supersede oral tradition by the creation of a literary ‘counterform’ see J. Halverson, ‘Oral and
Written Gospel: A Critique of Werner Kelber’, NTS 40 (1994) 180–95.
 Why were early Christians addicted to the codex? 187
private small rolls). They did this because otherwise they would have transgressed
the law.91
In making this suggestion Lieberman set a cat among the form-critical
pigeons – but the pigeons simply looked the other way. There were, how-
ever, two notable exceptions. In 1964 Birger Gerhardsson re-examined the
rabbinic evidence in the light of Lieberman’s comments. He confirmed that
in rabbinic circles written notes were used to facilitate the private repetition
and the maintenance of knowledge of the oral Torah. Teachers and pupils
used writing tablets and notebooks – and this may have been more usual
in Palestine than in Babylon, perhaps under the influence of Hellenistic
schools of rhetoric and philosophy.92 Gerhardsson’s book was misread and
criticized strongly in some circles, but a rehabilitation is now under way,
with the strong support of Jacob Neusner’s extended Preface to the 1998
second edition.93 His general point concerning the use of notebooks in some
rabbinic circles is based on solid textual evidence, especially the reference
in m. Kelim 24.7 to three kinds of writing tablets.94
C. H. Roberts and T. C. Skeat also drew on Lieberman’s comments
and appealed to m. Kelim 24.7: ‘three kinds of writings tablets, wax, ivory,
“and those of papyrus”’. ‘It is possible, therefore, that papyrus tablets were
used to record the Oral Law as pronounced by Jesus, and that these tablets
might have developed into the primitive form of codex.’95 This supposed
reference to papyrus notebooks became a linchpin in their preferred ex-
planation for the Christian adoption of the codex. However, Colette Sirat
has undermined their theory by showing that Lieberman misinterpreted
the rabbinic references to pinkasim: they were not made of papyrus, but of
wood.96
M. Kelim 24.7 suggests that it is highly likely that the first followers of
Jesus were familiar with notebooks made from wood, both wax stylus tablets

91 S. Lieberman, Hellenism in Jewish Palestine, 2nd edn (New York: Jewish Theological Seminary, 1962),
p. 203.
92 B. Gerhardsson, Memory and Manuscript: Oral Tradition and Written Transmission in Rabbinic
Judaism and Early Christianity (Lund: Gleerup, 1964), pp. 29, 157–63.
93 B. Gerhardsson, Memory and Manuscript, 2nd edn (Grand Rapids: Eerdmans, 1998), pp. xxv–xlvi.
94 Herbert Danby’s translation of m. Kelim 24.7 is as follows: ‘There are three kinds of writing-tablet:
that of papyrus, which is susceptible to midras-uncleannness; that which has a receptacle for wax,
which is susceptible to corpse-uncleanness; and that which is polished, which is not susceptible to
any uncleanness.’ The Mishnah (Oxford: Oxford University Press, 1933).
95 Roberts and Skeat, Birth, p. 59.
96 Colette Sirat, ‘Le Livre hébreu dans les premiers siècles de notre ère: le témoinage des textes’, in
J. Lemaire and E. van Balberghe, eds., Calames et cahiers: mélanges de codicologie et de paléographie
offerts à Léon Gilissen (Brussels: Centre d’étude des manuscripts, 1985), pp. 169–76; reprinted in Les
Débuts du codex, pp. 115–24.
188 The gospels and papyri codices
and ink leaf tablets, and perhaps ivory as well. As noted above, an example
of an ink leaf tablet has been found in a cave near the Dead Sea, and dated
to ad 134-5. I am not for a moment suggesting that Jesus traditions did
not circulate orally, but those traditions were first transmitted in Herodian
Palestine, and we now have a mountain of evidence gathered together by
A. R. Millard, which shows that writing was widespread at all levels of
society.97 Millard’s conclusion, based on extensive knowledge of writing in
the Near East generally, is surely sound: ‘far more weight than has been
allowed should be given to the role of writing in preserving information
about Jesus of Nazareth from his lifetime onwards and so in forming the
Gospel tradition’.98
There are important implications for the Q hypothesis which I accept
as the most plausible explanation of the extensive non-Marcan agreements
between Matthew and Luke. The consensus view at present is that Q was
a written document drawn on by both Matthew and Luke. But supporters
of this version of the Q hypothesis always shuffle when confronted with
the very considerable variation in the level of agreement between Matthew
and Luke in alleged Q passages. In some passages the level of agreement
is as high as 90 per cent, and even almost 100 per cent in shorter sections
(compare, for example, Matt. 11.2-27 and Luke 7.18-35; 10.12-15, 21-2). But
in other passages agreement is as low as 10 per cent (e.g. Matt. 25.14-30 and
Luke 19.12-27).99 If we take seriously the possibility that Jesus traditions
were transmitted both orally and in written form, then an explanation is to
hand. Q passages where there is close agreement may come from a written
document or from more than one set of notes; where the level of agreement
is low, oral traditions may have been used.100
Ulrich Luz and his pupil Migaku Sato have made a similar proposal.
Luz suggests that Q traditions may have been collected in a rather large
notebook ‘bound together with strings on the margin. It permitted an
insertion of new leaves at any time. The Gospel of Mark, however, was
a solidly bound codex and therefore a literary work which for this reason
continued to be handed down even after its expansion by Matthew.’101
The widely held view that the followers of Jesus were either illiterate
or deliberately spurned the use of notes and notebooks for recording and
97 Millard, Reading and Writing (above, n. 29), pp. 84–131. 98 Ibid., pp. 228–9.
99 See S. Hultgren, Narrative Elements in the Double Tradition (Berlin: de Gruyter, 2002), pp. 325–30
for a fresh discussion.
100 See now ibid., p. 328.
101 U. Luz, Matthew 1–7 (Edinburgh: T. & T. Clark, 1989), pp. 46–7. See also Migaku Sato, Q und
Prophetie. Studien zur Gattungs- und Traditionsgeschichte der Quelle Q, WUNT 2. Reihe, Vol. 29
(Tübingen: Mohr, 1988), pp. 72–7.
 Why were early Christians addicted to the codex? 189
transmitting Jesus traditions needs to be abandoned. Oral and written
traditions were not like oil and water.102 They could exist side by side;
orally transmitted traditions could be written down by the recipients – and
written traditions could be memorized and passed on orally.

8 .6 conclusions
The first followers of Jesus will have been familiar with several forms of
notebooks, whether they were in Judaea, in Galilee, or in eastern Mediter-
ranean cities. So it is natural to suppose that notebooks will have been
used for testimonia and for other collections of excerpts of Scriptural pas-
sages, for drafts and copies of letters, and for short collections of sayings of
Jesus, perhaps grouped thematically. A given Christian community may
have been familiar with all these forms of written traditions in notebook
format. But, once Christian scribes discovered how useful the ‘page’ format
was, it very quickly became the norm for copies of Paul’s letters and of the
gospels, and for Christian copies of the Scriptures.
When did Christian addiction to the codex format first begin to take
hold (my stage one)? In the past I have assumed, along with most scholars,
that this happened at the end of the first or early in the second century.
The corollary would then be that in the first post-Easter decades Christians
made considerable use of notebooks as sketched above, then rolls for both
letters and gospels written between, say, ad 50 and 90, before switching to
the codex early in the second century.
For two main reasons this scenario has seemed plausible to many. (1) The
first strong evidence for the use of the codex for writings more substantial
than notes or drafts emerges only towards the end of the first century
(Martial; P. Petaus 30), some decades after the first Christian writings began
to circulate. Christian scribes are more likely to have borrowed the new
format from their non-Christian counterparts than vice versa. (2) At least
some New Testament writings seem to have been designed for rolls. Luke’s
Gospel and his Acts would have fitted neatly into two ‘standard-sized’
rolls, and the prefaces to both writings strongly suggest that this was Luke’s
intention. Hebrews is a sophisticated writing, well suited to the roll format.
However, I am now much less confident about the consensus view. Once
big bang theories are abandoned, as I have argued above, a gradual transition

102 See J. Halverson’s vigorous protest against W. Kelber’s ‘almost metaphysical split between speech
and writing’. ‘Oral and Written Gospel’, 180–95.
190 The gospels and papyri codices
from notebook to codex is easier to envisage than the alternative, i.e. note-
book to roll, and then to codex very shortly afterwards.
Martial and the writers of P. Petaus 30 may well have had predecessors;
in neither case is there even a hint that the codex is a recent invention.
So use of the codex in the middle of the first century is perfectly possible.
Christian and non-Christian scribes may have begun to experiment with
the codex quite independently.
Perhaps the evangelist Mark did write his gospel in an early form of codex.
We would then have to reconsider the possibility that both the opening and
the ending of Mark’s Gospel have been lost, for this is much more likely
to have happened to an early version of the codex than to a roll.103 The
autograph of Luke’s more sophisticated ‘literary’ gospel may have been in
the roll format, but, by the time the first copies of Luke’s Gospel were made,
Christian addiction to the codex may have begun to take hold in earnest.
The speed with which the new format was adopted universally within
early Christianity is astonishing: I have noted above a set of factors which
encouraged and strengthened precocious use of the codex. The deployment
and development of nomina sacra, the method of abbreviation used for
‘God’, ‘Jesus’, and other related terms, is a related question which cannot
be discussed here.104 Taken together, addiction to the codex and addiction
to the use of nomina sacra suggest that Christian communities around the
eastern Mediterranean were in closer touch with one another than is usually
supposed to have been the case.
103 J. K. Elliott claims that Mark 1.1–3 contains too many non-Marcan features to stem from the
hand of the evangelist. He suggests that both the original opening and ending of Mark were lost
shortly after this gospel circulated in codex form at the beginning of the second century. The
present opening, Mark 1.1–3, and the present conclusion, Mark 16.9–20, were added in the second
century. Elliott notes that the outer leaves of a codex are particularly vulnerable to damage or loss.
He implies that Mark was written on a roll before being transferred to a codex. ‘Mark 1.1–3 – A
Later Addition to the Gospel’, NTS 46 (2000) 584–8. N. Clayton Croy’s stimulating book, The
Mutilation of Mark’s Gospel (Nashville: Abingdon, 2003), reached me while my typescript was in
the press. Croy argues that Mark may have existed in single-quire codex form ‘at a very early stage,
perhaps even in the autograph, and that damage to the autograph (or perhaps to an early copy
from which all manuscripts derive) resulted in simultaneous loss of the first and last leaves. It is
possible (although I think it less likely) that Mark was originally composed on a scroll and that it
lost both beginning and ending in separate accidents’ (p. 168; see also pp. 149–52). Croy’s cautious
conclusion is broadly in line with my own comments above; he implies that Martial’s codices may
have encouraged Christian scribes to use the codex format for the autograph and/or earliest copies
of Mark’s Gospel (p. 150).
104 See now C. M. Tuckett’s thorough discussion with full bibliographical references, ‘ “Nomina Sacra”:
Yes and No?’, in J.-M. Auwers and H. J. de Jonge, eds., The Biblical Canons (Leuven: Peeters, 2003),
pp. 431–58. See also A. R. Millard, ‘Ancient Abbreviations and the Nomina Sacra’, in C. Eyre, A.
Leahy, and L. M. Leahy, eds., The Unbroken Reed: Studies in the Culture and Heritage of Ancient
Egypt (London: The Egypt Exploration Society, 1994), pp. 221–6; L. W. Hurtado, ‘The Origin of
the Nomina Sacra: A Proposal’, JBL 117 (1998) 655–73.
 Why were early Christians addicted to the codex? 191
There is a further consideration which has a direct bearing on the early
reception of the gospels. We need not suppose that, once the codex began
to be used for early Christian writings, notebooks with Jesus traditions (or
with testimonia or collections of Scriptural passages) immediately ceased
to be used. Quintilian speaks of seeing Cicero’s own notes for some of
his speeches, which were still in circulation more than a century after the
author’s death!105 So why should we not suppose that notebooks with Jesus
traditions continued to be used even after copies of individual gospels began
to circulate? Christian communities which did not have a copy of a full
Gospel may have had to make do with parchment or papyrus notebooks for
some time. So Christian missionaries and teachers may well have continued
to use parchment or papyrus notebooks with Jesus traditions (and OT
passages) alongside copies of the gospels and, of course, oral traditions. If
so, we should not be surprised at the varied ways OT and Jesus traditions
are cited or alluded to in the apostolic fathers and in the writings of Justin
Martyr.106
This chapter opened with Sir Frederic Kenyon’s observation that the
Chester Beatty Biblical papyri revolutionize our knowledge of the early
history of the codex form of book. Perhaps we may dare to hope that further
discoveries of manuscripts as important as the Chester Beatty Biblical papyri
will extend our knowledge of the transition from the roll to the codex-book
format, and in particular of the reasons why Christian scribes were the first
to become addicted to the codex. For the time being we do have enough
new evidence to suggest that several widely held assumptions concerning
the transmission of traditions in early Christianity need to be reconsidered.

105 Quintilian, Inst. Or. x.7.30–1.

106 For further discussion, see section 2.9 of Chapter 2 and Chapter 4 above.
 chap t e r 9

What are the gospels? New evidence

from papyri?

In recent decades the question, ‘What are the gospels?’ has been discussed
from three perspectives. I shall refer to two briefly, before concentrating on
the third.
In discussion of this question, pride of place must always go the literary
genre of the gospels. What kind of writings are we dealing with? Histories,
religious novels, biographies, early Christian sermons in narrative dress,
or catechetical handbooks? The very first step in the interpretation of any
writing, whether ancient or modern, is to establish its literary genre.1 If we
make a mistake about the literary genre of the gospels, interpretation will
be skewed or even misguided. A decision about the genre of a work and
the discovery of its meaning are inextricably inter-related; different types
of text require different types of interpretation.2
In this chapter I do not propose to consider yet again whether or not the
gospels are biographies. I have had my say on that topic more than once.3
Following intensive recent discussion, broad agreement has been reached.
The gospels are now widely considered to be a sub-set of the broad ancient
literary genre of b©oi, biographies. Even if the evangelists were largely ig-
norant of the tradition of Greek and Roman b©oi, that is how the gospels
were received and listened to in the first decades after their composition.4
The question, ‘What are the gospels?’ has recently been given an un-
expected answer: they were intended to be writings for all Christians. For
1 Cf. E. D. Hirsch, Validity in Interpretation (New Haven: Yale, 1967), p. 76: ‘All understanding of
verbal meaning is necessarily genre-bound.’
2 On this see A. Fowler, Kinds of Literature (Oxford: Oxford University Press, 1982), p. 38; Hirsch,
Validity, p. 113.
3 I first tackled this topic when it was right out of fashion. See G. N. Stanton, Jesus of Nazareth in New
Testament Preaching (Cambridge: Cambridge University Press, 1974), pp. 117–36; more recently, The
Gospels and Jesus, 2nd edn (Oxford: Oxford University Press, 2002), pp. 13–18.
4 See especially R. A. Burridge, What are the Gospels? A Comparison with Graeco-Roman Biography
(Cambridge: Cambridge University Press, 1992; 2nd edn Grand Rapids: Eerdmans, 2004); D.
Frickenschmidt, Evangelium als Biographie. Die vier Evangelien im Rahmen antiker Erzählungskunst
(Tübingen and Basle: A. Francke, 1997).

192
 What are the gospels? New evidence from papyri? 193
decades now most writers on the gospels have accepted as axiomatic that
the four evangelists all wrote for their own particular Christian communi-
ties. However, Richard Bauckham and a number of British scholars have
offered a ‘new paradigm’:5 the evangelists did not write for one Christian
community, or even for a cluster of Christian communities, but for all
Christians everywhere. Widely held assumptions concerning the intentions
of the evangelists and the audiences they envisaged have been confronted
head on.
I welcome some of the essayists’ concerns, but not quite unreservedly.6
I have protested more than once against the failure to appreciate that the
gospels are not letters directed to the problems of a particular early Christian
community: the primary intention of the evangelists is to set out the story
of Jesus of Nazareth. In recent years I have become increasingly uneasy
about the constructs, ‘the Q community’, ‘the Matthean community’, ‘the
Johannine community’, and so on.7 However, firm new evidence for the
actual circulation of the first copies of the gospels is not set out in any of
the essays edited by Bauckham. Equally conspicuous by its absence is the
evidence of the earliest surviving copies of the gospels, evidence which will
concern us in this chapter.
I intend to take up the question, ‘What are the gospels?’ from a third
angle, which deserves more attention than it has received. I am convinced
that the early papyri of the gospels can make a contribution to this question.
I concede immediately that the contribution is likely to be limited, for
the earliest papyri have survived by chance in only one geographical area,
and thus they may not be representative of all the early copies of the
gospels. Nonetheless, here we have evidence which has grown significantly
in the last few years, and which has been overlooked all too often. In the
impressive array of major commentaries, as well as in the continuing flood
of monographs, next to nothing is said about the earliest surviving papyri,
even though their importance for the textual traditions of the gospels can
hardly be overestimated.
In this chapter I shall discuss the earliest papyri fragments of the gospels,
with special reference to the Oxyrhynchus papyri published between 1997

5 R. J. Bauckham, ed., The Gospels for All Christians: Rethinking the Gospel Audiences (Grand Rapids:
Eerdmans, 1998).
6 See P. F. Esler, ‘Community and Gospel in Early Christianity: A Response to Richard Bauckham’,
SJT (1998) 235–48 and R. Bauckham’s reply in the same issue (pp. 249–53); David Sim, ‘The Gospels
for all Christians? A Response to Richard Bauckham’, JSNT 84 (2001) 3–27.
7 See G. N. Stanton, ‘Revisiting Matthew’s Communities,’ in SBL Seminar Papers 1994, ed. E. H.
Lovering Jr. (Atlanta: Scholars Press, 1994), pp. 9–23.
194 The gospels and papyri codices
and 1999.8 These papyri force us to reconsider an influential answer to the
question: what are the gospels? On this view, a sharp contrast is drawn
between the carefully produced contemporary copies of Jewish writings on
rolls and early copies of the gospels in codices. Whereas Jewish rolls are said
to have been written with care in a formal hand, the gospels are said to be
the ‘workaday’, ‘utilitarian’, ‘downmarket’ handbooks of an inward-looking
sect. I hope to show that this differentiation is unwarranted.

9. 1 the gospels as utilitarian handbooks

Several Greek palaeographers have been more astute than New Testament
specialists in their use of the evidence of the earliest copies of the gospels.9
In his important Schweich lectures, Manuscript, Society and Belief in Early
Christian Egypt, C. H. Roberts broke new ground.10 As his title implies,
Roberts shows how early Christian manuscripts shed light on the social and
religious setting of early Christianity in Egypt. I am convinced that Roberts
was asking the right questions, but the recently published papyri force us
to modify several of his conclusions – conclusions which have influenced
a number of New Testament scholars.
First I must sketch the consensus view, of which the palaeographers C. H.
Roberts and Sir Eric Turner11 are the most distinguished proponents.12 In
their view, manuscripts written in Greek in the early centuries of Christian
era can be divided into two groups: ‘bookhands’ and ‘documentary hands’.
Bookhands were usually used for literary works, but occasionally for other
8 E. W. Handley, U. Wartenberg, et al., eds., The Oxyrhynchus Papyri, Vol. lxiv (London: Egypt
Exploration Society for the British Academy, 1997); M. W. Haslam et al., eds., The Oxyrhynchus
Papyri, Vol. lxv (1998); N. Gomis et al., eds., The Oxyrhynchus Papyri, Vol. lxvi (1999). For discussion
of the textual evidence of these papyri, see J. K. Elliott, ‘Five New Papyri of the New Testament’, NovT
41 (1999) 209–13, and ‘Seven Recently Published New Testament Fragments from Oxyrhynchus’,
NovT 42 (2000) 209–13; P. M. Head, ‘Some Recently Published NT Papyri from Oxyrhynchus: An
Overview and Preliminary Assessment’, TB 51 (2000) 1–16.
9 This section and the following one have been adapted and revised from part of my ‘The Early
Reception of Matthew’s Gospel’, in D. E. Aune, ed., The Gospel of Matthew in Current Study (Grand
Rapids: Eerdmans, 2001), pp. 42–52.
10 C. H. Roberts, Manuscript, Society and Belief in Early Christian Egypt (London: Oxford University
Press for the British Academy, 1979).
11 E. G. Turner, The Typology of the Early Codex (Philadelphia: University of Pennsylvania Press, 1977).
See also E. G. Turner, Greek Manuscripts of the Ancient World (Oxford: Oxford University Press,
1971; 2nd edn, 1987); G. Cavallo and H. Maehler, Greek Bookhands of the Early Byzantine Period,
ad 300–800, Bulletin Supplement 47 (London: University of London Institute of Classical Studies,
1987).
12 I cannot claim to be a palaeographer myself, and stand in awe of the specialists whose publications
I have consulted. I have looked carefully at the plates of the papyri discussed below: the general
characteristics of the styles of handwriting used seem to me to be clear. However, on palaeographical
questions, I have quoted the judgements of the specialists, not my own.
 What are the gospels? New evidence from papyri? 195
kinds of writings. The manuscripts were carefully written by very skilled
scribes, often by slaves. Bookhands use upright, independent letters (not
ligatures), often with serifs; they are usually bilinear, i.e. individual letters
rarely protrude beyond two imaginary horizontal lines on the papyrus or
parchment. Cursive forms are not used. Manuscripts prepared carefully in
this way were more expensive; they were used by the more educated elite.
The ‘classic’ examples of Christian manuscripts in this style are Codex
Sinaiticus and Codex Vaticanus from the fourth century. Sinaiticus is a
splendid ‘pulpit edition’ of the whole Greek Bible; Vaticanus is a ‘pulpit
edition’ of the gospels, Acts, and most of the Pauline corpus.13
Documentary hands were used for workaday documents of all kinds.
They were written rapidly, usually with long lines, and with little concern
for bilinearity. Some individual letters were joined together (i.e. ligatures
were often used). Manuscripts written in this way were often used in a
business setting, or for private use. In comparison with ‘bookhands’, they
were ‘downmarket’.
However, it is important to note that there are literary texts which
are written ‘cursively’ in documentary hands, and there are documents in
which the letters are generally formed separately, with bilinearity. A copy
of Aristotle’s ìAqhna©wn Pol©teia (On the Constitution of Athens) on a roll
from the end of the first century is a good example of the former. It has
obviously been written very quickly, with little regard for the appearance of
the text. In places, up to five letters are joined together in a single sequence.
It is written on the back of four rolls, the rectos of which have agricultural
accounts dated ad 78/9. Presumably this was a private rather than a library
copy of Aristotle. A private letter dated to c. ad 170, P. Oxy. 2192, is a good
example of a document written in a bookhand in a competent professional
manner by a skilled scribe. A hastily written postscript in documentary
cursive style has been added by the author, who had presumably dictated
the letter.14
In his Schweich lectures, C. H. Roberts contrasted sharply the ‘hieratic
elegance of the Graeco-Jewish rolls of the Law’ with the ‘workaday appear-
ance of the first Christian codices (whether of OT or Christian writings)’
(p. 19). With a few exceptions, Christian manuscripts are ‘based on . . . the
model of documents, not that of the Greek classical manuscripts nor on that
of the Graeco-Jewish tradition’ (p. 20). Roberts concluded that, if Christian
13 For an important fresh discussion of the origin of Sinaiticus and Vaticanus, see T. C. Skeat, ‘The
Codex Sinaiticus, the Codex Vaticanus, and Constantine’, JTS 50 (2000) 583–625.
14 For the two examples, compare Plates 60 and 68 in E. G. Turner, Greek Manuscripts, 2nd edn, pp. 102
and 114.
196 The gospels and papyri codices
communities are faintly mirrored in their books, they would seem ‘to have
been composed not so much of intellectuals or the wealthy as of ordinary
men of middle or lower classes’ (p. 25).
In 1977 Sir Eric Turner made similar observations in his now classic study
of papyrus codices of the second or third centuries. He noted that it is not
easy to find examples of calligraphy among papyrus codices of this period.
‘Their handwriting is in fact often of an informal and workaday type,
fairly quickly written, serviceable rather than beautiful, of value to a man
interested in the content of what he is reading rather than its presentation.’
Turner then lists some examples, including the majority of the Chester
Beatty codices. ‘These give the impression of being “utility” books; margins
are small, lines usually long’; their status is second-class in comparison with
the contemporary papyrus roll.15
The views of these two giants among palaeographers have been echoed
by several New Testament scholars. In 1997 Harry Gamble noted that a
bookhand is rarely found in Christian texts before the fourth century; with
Sinaiticus and Vaticanus ‘a barrier was broken’: ‘never before had Christian
books been so fine’.16 Until that time, Christians had little aesthetic regard
for their literature, and they do not seem to have had a cultic attitude to-
wards their books. In contrast, Jewish Greek manuscripts are better written:
they ‘usually display an even, formal script with tendencies not only towards
a bookhand but toward a somewhat decorated style with footed and serifed
letters’.17
Similar comments have been made recently by Loveday Alexander18 and
by Bart Ehrman.19 Hence it is not an exaggeration to claim that the C. H.
Roberts–E. G. Turner view has become the consensus. It will be apparent
that it undermines, at least in part, the contention that the gospels are a sub-
set of Graeco-Roman biography, for the latter was not normally second-
class and downmarket. And it places an even firmer question mark against
various literary approaches to the gospels which have become fashionable in
recent decades. Are the evangelists likely to have been rather sophisticated
writers who used a whole gamut of literary techniques, if the earliest copies

15 Turner, Typology, p. 37.

16 Harry Y. Gamble, Books and Readers in the Early Church: A History of Early Christian Texts (New
Haven and London: Yale University Press, 1995), p. 80.
17 Ibid., pp. 78–9.
18 Loveday Alexander, ‘Ancient Book Production and the Circulation of the Gospels’, in Bauckham,
The Gospels for All Christians, pp. 71–111.
19 Bart Ehrman, ‘The Text as Window: New Testament Manuscripts and the Social History of Early
Christianity’, in B. D. Ehrman and M. W. Holmes, eds., The Text of the New Testament in Contem-
porary Research (Grand Rapids: Eerdmans, 1995), pp. 372–5.
 What are the gospels? New evidence from papyri? 197
of their work were decidedly workaday and utilitarian? Were the evangelists’
literary skills not appreciated by early copyists? Or were early Christian
communities not able to afford to produce copies of their writings at the
level of sophistication they deserved?
We do not need to trouble ourselves with such questions, for the recently
published papyri force us to modify the consensus view. I shall set out below
basic details of what are almost certainly the earliest surviving papyri of the
four gospels.20 The papyri are all from no later than the middle of the
third century, though papyrologists are always rightly wary about dating
precisely what are often very fragmentary writings. I shall set out the papyri
according to their ∏ number (as used in modern editions of the Greek
text); this corresponds closely to the order in which the papyri were first
published.21

9.2 early papyri of mat t hew

We now have seven papyri of Matthew which were probably written before
the middle of the third century.
(1) ∏1 (= P. Oxy. 2; 1907). This small fragment of Matthew’s genealogy
(Matt. 1:1-9; 14-20) is usually dated to the early third century. The hand is
generally bilinear, with only a few ligatures. It is at least as close to a book-
hand as to a documentary hand. When this fragment was first published
in 1898, it was considered to be ‘the oldest known manuscript of any part
of the NT’.22
(2) ∏45 (= P. Chester Beatty I; 1933). This codex, which is usually dated
to the first half of the third century, contains parts of the final chapters of
Matthew (Matt. 20:24-32; 21:13-19; 25:41-26:39), as well as parts of Mark,
Luke, John, and Acts. The first editor noted that the codex is written by a
competent scribe, but without calligraphic pretensions, in a small and very
clear hand; the letters have a decided slope to the right, as opposed to the
uprightness generally found in Roman hands of the first two centuries.23

20 For full details of the original publication of the fragments, photographic plates, and bibliography, see
J. K. Elliott, A Bibliography of Greek New Testament Manuscripts (Cambridge: Cambridge University
Press, 1989; 2nd edn, 2000). See also K. Aland, ed., Kurzgefasste Liste der griechischen Handschriften
des Neuen Testaments, 2nd edn (Berlin: de Gruyter, 1994).
21 For fuller details of (1)–(5) in my list, see K. Aland, Repertorium der griechischen christlichen Papyri,
Vol. i: Biblische Papyri (Berlin and New York: De Gruyter, 1976).
22 Oxyrhynchus Papyri, ed. Bernard P. Grenfell and Arthur S. Hunt, Vol. i (London: Egypt Exploration
Fund, 1898), p. 4. There is a good reproduction of this fragment as the frontispiece to this edition.
23 F. G. Kenyon, ed., The Chester Beatty Biblical Papyri, Vol. ii (London: Emery Walker, 1933),
pp. viii–ix.
198 The gospels and papyri codices
T. C. Skeat, the doyen of contemporary papyrologists, has recently observed
that, in contrast to ∏64 + ∏67 + ∏4 ((4) below in this list of early Matthean
papyri), ∏45 was ‘not intended for liturgical use’.24 In the opening pages of
Chapter 8 above, attention was drawn to the significance of the publication
of the Chester Beatty Biblical papyri by Sir Frederic Kenyon for the origin
of the codex. However, it also has to be said that ∏45 has exercised too great
an influence on studies of the handwriting of the early papyri. As we shall
see, the hand of ∏45 is not the norm.
(3) ∏53 (Ann Arbor, University of Michigan, Inv. 6652; 1937). This small
fragment of parts of Matt. 26:29-40 may date from the middle of the third
century. The hand is described as ‘semi-uncial’, with upright symmetrical
letters.25
(4) ∏64 (Magdalen College, Oxford, Gr. 18; 1957) + ∏67 (P. Barcelona
I; 1961). These fragments of Matthew 3, 5, and 26 are from the same codex
as the more extensive fragments of Luke 1, 3, and 5 known as ∏4 (Paris,
Bibliothèque nationale, Suppl. Gr. 1120, 1892). ∏64 was considered by its
first editor, C. H. Roberts, to date from the end of the second century.
In spite of the attempt of C. P. Thiede to date ∏64 and ∏67 to the first
century, Roberts’s dating is still generally accepted.26 T. C. Skeat’s theory
that ∏64 + ∏67 + ∏4 are from the same four-gospel codex is winning
wide support.27
C. H. Roberts recognized that ∏64 was an early example of a Biblical
uncial hand: ‘a thorough-going literary production’, i.e. in effect a prede-
cessor of Codex Sinaiticus and Codex Vaticanus.28 In a letter to me dated
30 March 1996 Skeat echoed Roberts’s comment, and noted that the hand
of ∏64 + ∏67 + ∏4 is quite clearly a literary one.
The significance of one of the codex’s most striking features, its double
columns, has not received adequate attention. The format of two columns
to the page is rare in papyrus codices. This is the only example of a two-
column Greek New Testament papyrus manuscript, though there are four

24 T. C. Skeat, ‘The Oldest Manuscript of the Four Gospels?’, NTS 43 (1997) 1–34. See also pp. 71–5
above.
25 See Aland, Reportorium, p. 283. A separate sheet, probably not from the same codex, contains parts
of Acts 9.33–10.1.
26 For full discussion and references, see G. N. Stanton, Gospel Truth? New Light on Jesus and the Gospels
(London: HarperCollins, 1995), pp. 1–10. The 1997 paperback edition includes an Afterword with
further references. For discussion of Thiede’s rash theory, see the references given above on p. 73
n. 37.
27 Skeat, ‘The Oldest Manuscript’, pp. 1–34.
28 Roberts, Manuscript, p. 13. Skeat noted that in this codex ‘organised text-division is now carried back
well into the second century’. ‘The Oldest Manuscript’, p. 7.
 What are the gospels? New evidence from papyri? 199
examples in early fragments of OT papyri.29 The narrow columns, with
only about fifteen letters in each column, would have assisted reading aloud
in the context of worship. So the use of two columns in ∏64 + ∏67 + ∏4
is almost certainly an indication of a high-class codex, a splendid ‘pulpit
edition’ intended for liturgical use.30
There are several other indications that this codex was an édition de luxe.31
The codex was planned and executed meticulously: the skill of the scribe
in constructing it is most impressive. All these features indicate a most
handsome edition of the four gospels, which would have been expensive to
produce.
(5) ∏77 (= P. Oxy. 2683; 1968). This small fragment contains parts of
Matt. 23.30-9. Another fragment from the same page of Matthew 23 has
recently been published (P. Oxy. lxiv 4405 (1997)). The first editor of
P. Oxy. 2683, Peter Parsons, noted that ∏77 is ‘delicately executed with a
fine pen’, to be dated to the late second century, and hence among the
oldest New Testament texts.32 C. H. Roberts noted its elegant hand, and
its use of ‘what was or became a standard system of chapter division as well
as punctuation and breathings’.33
(6) ∏103 (= P. Oxy. 4403; 1997). This small fragment of Matthew 13 and
14 is dated by its editor, J. D. Thomas, to the late second or the early third
century.34 In Thomas’s opinion ‘the hand is quite elegant, with noticeable
hooks at the top of most hastas, and occasional serifs elsewhere’. The hand
is so similar to P. Oxy. 4405 = ∏77 ((5) above) that Thomas allows the
possibility that they may be from the same codex, though he concludes
that it is ‘safest to treat the papyri as from two different codices’.
(7) ∏104 (= P. Oxy. 4404; 1997). J. D. Thomas considers that we may
assign this fragment of Matthew 21 ‘with some confidence to the second half
of the second century, while not wishing to exclude altogether a slightly
earlier or a slightly later date’.35 This cautious judgement is a mark not
of indecisiveness, but of the difficulty papyrologists have in dating small
fragments. Thomas notes that the hand is very carefully written, with

29 See E. G. Turner’s list of papyrus codices written in two columns, Typology, p. 36.
30 Cf. ibid., pp. 36–7. J. van Haelst, Catalogue des papyrus littéraires juifs et chrétiens, Papyrologie I (Paris:
Publications de la Sorbonne, 1976), no. 336, even says that this is the oldest example of a codex with
two columns; presumably he means the oldest Biblical codex, though, in the light of Turner’s list,
this is a doubtful claim.
31 This is T. C. Skeat’s phrase, ‘The Oldest Manuscript’, p. 26.
32 P. J. Parsons, in The Oxyrhynchus Papyri, Vol. xxxiv (1968).
33 Roberts, Manuscript, p. 23.
34 J. D. Thomas, in Handley et al., The Oxyrhynchus Papyri, Vol. lxiv, pp. 5–6.
35 J. D. Thomas, in The Oxyrhynchus Papyri, Vol. lxiv, p. 8.
200 The gospels and papyri codices
extensive use of serifs; bilinearity is strictly observed. Once again Thomas
refers to the hand as ‘elegant’.

In formulating what I have dubbed above ‘the consensus view’, C. H.

Roberts drew on a list of what he considered in 1970 to be the ten ear-
liest Christian Biblical papyri and the four earliest Christian non-Biblical
papyri.36 Roberts emphasized that, with three exceptions, ‘there is no calli-
graphic hand in the group’.37 In ten cases out of fourteen, Roberts detected
‘basically a documentary hand’; these earliest papyri might be described
as ‘reformed documentary’.38 The three exceptions were (4) and (5) in my
list above, along with P. Oxy. I, 1, one of the three Greek fragments of the
Gospel of Thomas.
In the list above of the seven earliest papyri of Matthew’s Gospel, (4) and
(5) are not, as C. H. Roberts supposed, exceptions to the general pattern of
‘reformed documentary’ hands: they are the rule! The recent publication
of (6) and (7) has altered the picture very considerably: four of the earliest
seven papyri of Matthew are at least as close to ‘bookhands’ as they are to
‘reformed documentary’ hands. It is worth noting that, even though dating
papyri is a skilled art rather than a science, the more literary hands among
the seven (my (4), (5), (6), and (7)) have the strongest claims to be the
earliest of the group.
The fact that four of the seven earliest surviving papyri of Matthew are
carefully written in an elegant hand is significant, but it is of course possible
that future publication of further papyri may spring further surprises. My
main point is that the consensus view now needs to be modified: the earliest
surviving copies of the gospels present a much more diverse picture than
recent scholarship has allowed. The seven earliest papyri of Matthew suggest
that this gospel was used both in private and in public settings. Matthew’s
Gospel was not considered by some of those who copied it and used it in
the second half of the second century to be of ‘second-class’ status, quite
without literary pretensions.

9.3 early papyri of john

The provisional conclusions set out in the preceding three paragraphs now
need to be tested against the evidence of the early papyri of John’s Gospel.
36 C. H. Roberts, ‘Books in the Graeco-Roman World’, in The Cambridge History of the Bible, ed.
P. R. Ackroyd and C. F. Evans (Cambridge: Cambridge University Press, 1970), pp. 12–23.
37 Roberts, Manuscript, pp. 12–13.
38 Ibid., p. 14.
 What are the gospels? New evidence from papyri? 201
At present there are ten papyri of John’s Gospel which seem to date from no
later than the middle of the third century. Four of the ten were published as
recently as 1998.39 This time there is no need to comment on the papyri one
by one: all the papyri of John’s Gospel published up to 1995 have been edited
carefully and republished with full transcriptions and extensive plates.40
Of the ten early papyri of John, four may be said to have been written
in a bookhand or a ‘near’ bookhand: ∏66, ∏90, ∏95, and ∏108. Three
are perhaps closer to a bookhand than to a documentary hand: ∏52, ∏75,
and ∏109. The three remaining early papyri are probably closer to a docu-
mentary hand than to a bookhand, though none is a classic example of a
documentary hand: ∏45, ∏106, ∏107. In short, the early papyri of John are
written in a range of hands from ‘near’ bookhands to ‘reformed’ documen-
tary hands. As we noted above, the latter term was used by C. H. Roberts
to categorize the style of handwriting used as the norm in early Christian
papyri.
I shall now comment briefly on the four at the upper end of the spectrum,
for they most clearly undermine what I have called the current consensus,
i.e. the view that the gospels were at first considered to be utilitarian,
‘second-class’ writings using a ‘reformed documentary’ hand.
(1) ∏66 (= P. Bodmer II; c. ad 200; 1956). In his initial edition of
this manuscript Victor Martin included extended comments on the fine
hand used. He described it as a strongly stylized hand, which could even be
described as ‘literary’. He noted that the letters are strictly independent and
largely bilinear, and even noted some similarities with the classic bookhand
of Codex Sinaiticus.41
(2) ∏90 (= P. Oxy. 3523; 1983). These fragments of John 18 and 19
are dated confidently to the end of the second century.42 The well-formed,
decorated style with some small serifs is basically bilinear. It is considered to
be similar to ∏104, (7) in the preceding list of early Matthean papyri, which
is described as an ‘elegant hand’.43 The editor of ∏90, T. C. Skeat, noted

39 See Vol. lxv of The Oxyrhynchus Papyri. All four papyri, P. Oxy. 4445–8 (= ∏106–9), have been
edited by W. E. H. Cockle.
40 See W. J. Elliott and D. C. Parker, eds., The New Testament in Greek, Vol. iv, The Gospel According
to St John: The Papyri (Leiden: Brill, 1995). Four further papyri are listed in Nestle–Aland, 27th edn,
as third century: ∏5, ∏22, ∏28, ∏39. But, as the specialists whose views are noted by Elliott and
Parker suggest late third century or even early fourth century, I have not included them here.
41 See V. Martin, Papyrus Bodmer II, évangile de Jean, chs. 1–14 (Cologne and Geneva: Bibliotheca
Bodmeriana, 1956), pp. 14–17.
42 See Elliott and Parker, The New Testament in Greek, Vol. iv, for details and plates (47a and b). T. C.
Skeat, in The Oxyrhynchus Papyri, Vol. l, ed. A. K. Bowman (1983), pp. 3–8, notes that the codex
may possibly have contained two gospels.
43 J. D. Thomas, in The Oxyrhynchus Papyri, Vol. lxiv (1997), p. 8.
202 The gospels and papyri codices
that the manuscript had been prepared carefully with blank space before a
change of speaker and that the original codex may well have contained two
gospels.
(3) ∏95 (= Florence, P. Laur. Inv. II/31; 1985), an early third-century
small fragment of parts of nine lines of John 5, is in the Biblical uncial or
majuscule style, i.e. not unlike (4) in my list of Matthean papyri.44 This
similarity can be seen clearly even by an untrained eye.
(4) ∏108 (= P. Oxy. 4447; 1998). Two joining fragments of the end of
John 17 and the beginning of John 18 are written in independently formed
letters, i.e. without ligatures. The editor, W. E. H. Cockle, notes that this
is a practised hand, a ‘handsome, medium-sized, upright, capital hand’,
‘firmly bilinear’.45 As metallic ink is used, the piece is unlikely to be earlier
than the first decades of the third century.

9. 4 early papyri of mark’s and luke’s gospels

We have only three copies of Mark’s Gospel on papyrus. ∏84, with parts
of Mark 2 and 6 (and John 5 and 17) dates from the sixth century. ∏88,
from the fourth century, contains almost the whole of Mark 2. Only ∏45
is probably from before the mid third century. This is the Chester Beatty
codex with parts of all four gospels and Acts, which was discussed above.
Alas, there are no fragments of Mark’s Gospel among the papyri published
in recent years.
If, as claimed by José O’Callaghan in 1972 and Carsten Thiede in several
publications since 1992, a fragment of a roll discovered in Cave 7 at Qumran
does contain parts of Mark 6:52-3, our knowledge of the text of this gospel
would be taken back 200 years.46 Several further questions concerning the
early reception of Mark would arise. Since the roll was probably placed in
Cave 7 shortly before the Romans arrived in ad 68, Mark must have been
written earlier than is generally accepted. If Mark was written in Rome
(as most scholars hold), how did it arrive at Qumran very shortly after its
composition? Is it conceivable that Mark would have been of interest to
members of the Qumran community? And if not, are we to suppose that
someone else placed the roll in Cave 7? The questions multiply. Much is at
44 S. R. Pickering, Recently Published New Testament Papyri: P89–P95, Papyrology and Historical Perspec-
tives 2 (Sydney: The Ancient History Documentary Research Centre, Macquarie University, 1991),
p. 49. See Elliott and Parker, The New Testament in Greek, Vol. iv, for details and plates (48b and c).
45 P. Oxy. 4447 is edited by Cockle in The Oxyrhynchus Papyri, Vol. lxv, pp. 16–20.
46 See Carsten Thiede, The Earliest Gospel Manuscript? (Exeter, Paternoster, 1992); Carsten Thiede and
Matthew d’Ancona, The Jesus Papyrus (London: Weidenfeld & Nicolson, 1996). The latter book was
published in New York by Doubleday under the title Eyewitness to Jesus.
 What are the gospels? New evidence from papyri? 203
stake, so it is not surprising that discussion of this theory has been vigorous
over the last decade.
However, one implication of the theory that 7Q5 is a fragment of Mark’s
Gospel has received barely a mention in the literature. 7Q5, like all the
other Qumran writings, is written on only one side, so it is from a roll –
not a codex. If the fragment is from Mark, then this gospel was probably
first written on a roll, and only later were copies made in codices. Parts of
Chapter 8 of this book would have to be rewritten!
The claim that 7Q5 is a fragment of Mark 6 has been subjected to close
cross-examination.47 The jury has now returned. The verdict is unanimous.
7Q5 is not part of Mark’s Gospel! I shall be surprised if we hear anything
more of this theory.
At present we have only five papyri of Luke’s Gospel from before the
middle of the third century. (1) ∏4, from the end of the second century, is
from the same codex as ∏64 and ∏67, fragments of Matthew’s Gospel. I
noted above (pp. 198–9) C. H. Roberts’s view that it is ‘a thorough-going
literary production’. (2) ∏45, the Chester Beatty codex discussed above
(pp. 197–8), contains substantial parts of the central chapters of Luke: it
is said to be ‘without calligraphic pretensions’. (3) ∏69 (= P. Oxy. 2383;
1957) contains fragments of Luke 22 from a mid-third-century codex. Its
handwriting is said by the editor, E. G. Turner, to be ‘of the type found in
a good documentary rather than a literary hand’.48 (4) ∏75, the Bodmer
papyrus of Luke and John from early in the third century, was noted above
as probably closer to a bookhand than to a documentary hand. (5) ∏111
(= P. Oxy. 4495). These tiny fragments of five verses in Luke 17 were
published as recently as 1999 and dated to the first half of the third century.
The hand is said by the editor, W. E. H. Cockle, to be ‘semi-documentary’.49
Although it would be rash to draw far-reaching conclusions on the basis
of very limited evidence, the five early papyri of Luke turn out to have been
written in quite a wide spectrum of hands, as is more clearly the case with
the early papyri of both Matthew and John.

9. 5 early papyri and the reception of the gospels

We have far fewer early papyri of Mark and Luke than of Matthew and
John. And, if we were to consider all the papyri of the gospels, this would
47 See especially R. Gundry, ‘No NU in Line 2 of 7Q5: A Final Disidentification of 7Q5 with Mark
6:52-3’, JBL 118 (1999) 698–707.
48 E. Lobel et al., eds., The Oxyrhynchus Papyri, Vol. xxiv (1957), p. 1.
49 The Oxyrhynchus Papyri, Vol. lxvi.
204 The gospels and papyri codices
also be the case. While the sheer chance of discovery cannot be ignored,
the numbers are almost certainly significant: we may reasonably conclude
that Matthew’s and John’s Gospels were copied far more frequently than
Mark’s and Luke’s. This is very much in keeping with the evidence we have
of early quotations from, and allusions to, the gospels. The main reason
for the greater popularity of Matthew and John is not hard to find: they
were considered to have been written by the apostles themselves, and not,
as with Mark and Luke, by associates or later followers of apostles.
The recently published early Oxyrhynchus papyri fragments of Matthew
and John strongly suggest that these gospels were particularly popular in
Egypt, so local factors may have been at work. However, taken as a whole,
the Oxyrhynchus papyri do not simply reflect the preferences of one geo-
graphical area, for they were not all written in Egypt. Some are in Latin,
which would have been known by few in the town of Oxyrhynchus, where
the papyri were discovered. For example, P. Oxy. 30, a codex in Latin now
usually entitled De Bellis Macedonicis, probably originated in Rome.50 A
fragment of Irenaeus, P. Oxy. 405 (a roll), travelled all the way from Lyons to
Oxyrhynchus within twenty years of its production, ‘not long after the ink
was dry on the author’s manuscript’, to quote again Roberts’s memorable
comment.51
Several further points need to be made. We have noted that in the latter
half of the second century some copies of Matthew and John were made
with great skill and at some expense. The communities for which they were
made must have been reasonably wealthy, perhaps with a higher degree
of literacy than has usually been supposed.52 The carefully made copies,
some with signs of punctuation, strongly suggest that they were used for
liturgical reading in public rather than for private reading. They confirm
Justin Martyr’s comment from the middle of the second century that ‘the
memoirs of the apostles’, i.e. the gospels, were read in the context of worship
‘as long as time allows’ (I Apol. 67).
In Chapter 4 above, I noted that Justin’s phrase ‘the memoirs of the
apostles’ suggested that Justin considered the gospels to be literary writings
of some sophistication. So it should not surprise us that some of the earliest
papyri of the gospels, perhaps copied soon after Justin wrote, are ‘upmarket’.
50 Joseph van Haelst, ‘Les Origines du codex’, in Les Débuts du codex, ed. Alain Blanchard, Bibliologia
9 (Brepols: Turnhout, 1989), p. 27.
51 Roberts, Manuscript, p. 53. See also Chapter 3, p. 74 and n. 45.
52 E. J. Epp, ‘The Codex and Literacy in Early Christianity and at Oxyrhynchus: Issues Raised by Harry
Y. Gamble’s Books and Readers in the Early Church,’ Critical Review of Books in Religion 10 (1997)
29–32 suggests that the papyri from Oxyrhynchus encourage at least a questioning of the 10–20 per
cent average literacy rate for the entire Graeco-Roman world proposed by W. Harris.
 What are the gospels? New evidence from papyri? 205
They provide general support for my insistence in Chapter 4 that Justin
comes within a whisker of regarding sayings of Jesus and the gospels as
Scripture.
At the beginning of this chapter, I noted that both C. H. Roberts
and Harry Gamble have drawn a sharp contrast between the ‘hieratic
elegance’ of Graeco-Jewish writings ‘with their even, formal script with
tendencies not only towards a bookhand but toward a somewhat decorated
style with footed and serifed letters’ and the ‘workaday appearance’ of first
Christian codices.53 The recently published evidence suggests that Roberts
and Gamble were incorrect to differentiate Christian and Jewish attitudes
to authoritative texts on the basis of the handwriting styles used. The most
striking difference is the marked preference of early Christian scribes for
the codex format and their use of nomina sacra, the distinctive Christian
way of abbreviating ‘divine names’.54
This conclusion is strongly supported by two of the Oxyrhynchus papyri
published in 1998. P. Oxy. 4442 is a fragment of an early third-century codex
of a Septuagintal text of parts of Exodus 20; it has a good-sized formal
Biblical majuscule hand, basically bilinear, and much closer to a bookhand
than to a documentary hand. It is broadly comparable with ∏64 + ∏67
(Matthew) and ∏4 (Luke), (4) above in my list of early Matthean papyri.
There are two examples of the nomen sacrum qC in lines 11 and 16. The use
of the codex format and the standard Christian abbreviation of qe»v, ‘God’,
confirms that this is a Christian copy of Exodus. Unless Christian scribes
were reusing a roll by writing on the reverse side (and we have only four
examples), in copying Biblical writings they always used the codex format.
In contrast, P. Oxy. 4443 is a late first- or early-second century roll with a
Septuagintal text of parts of Esther 8 and 9, and of the fifth of the so-called
‘Additions to Esther’. It has frequent ligatures and cursive forms. Its editor,
K. Luchner, notes that it perhaps owes more to official documentary styles
than to bookhands. In line 12 there is no abbreviation of qeoÓ, confirmation,
along with the roll format, that it is Jewish rather than Christian.
There is no reason to doubt that the gospels were often used also as
‘workaday’, ‘utilitarian’ handbooks, or even as ‘school handbooks’ or ‘manu-
als of the teaching traditions of this pragmatically oriented group’.55 But

53 Roberts, Manuscript, pp. 19–20; Gamble, Books and Readers, p. 79.

54 On the latter, see especially C. M. Tuckett’s thorough discussion with full bibliographical references,
‘ “Nomina Sacra”: Yes and No?’, in J.-M. Auwers and H. J. de Jonge, eds., The Biblical Canons (Leuven:
Peeters, 2003), pp. 431–58; and L.W. Hurtado, ‘The Origin of the Nomina Sacra: A Proposal’, JBL
117 (1998) 655–73.
55 Alexander, ‘Ancient Book Production’ (above, n. 18), p. 105.
206 The gospels and papyri codices
the recently published papyri suggest that, by the second half of the second
century, much earlier than has usually been assumed, their literary qualities
and their authoritative status for the life and faith of the church were widely
recognized. By then, if not even earlier, some copies of the gospels were
prepared most carefully, probably for use in worship. The often-repeated
claim that the gospels were considered at first to be utilitarian handbooks
needs to be modified.
Early Christian obsession with the codex does not call these conclusions
into question. As we saw in Chapter 8, the codex was a natural development
from the widespread use of papyrus, parchment, and wooden notebooks.
But we should not suppose that the contents of the codex were necessarily
considered to be ‘second-class’. Martial confirms that towards the end of
first century the codex was used by travellers for the writings of Homer,
Virgil, Cicero and others; our earliest non-Christian papyrus codex is a
historical work, and we do have a handful of second-century codices with
literary texts.
Christians had their own good reasons for preferring the codex. What
continues to surprise is the almost universal adoption of this format by
Christian scribes, as well as their use of nomina sacra. That both ‘fashions’
caught on so quickly and so universally suggests that Christian communities
were in much closer touch with one another than we have usually supposed.
At the beginning of this chapter I welcomed C. H. Roberts’s insight in
1979 that early Christian papyri shed light on the social and religious setting
of early Christianity in Egypt. However, my conclusions differ. The more
recently published papyri suggest greater diversity than Roberts allowed.
In particular, some of the very earliest papyri of the gospels are far from
being fragments from ‘utilitarian handbooks’. ∏52, still considered to be
the very earliest papyrus fragment and thus the earliest material evidence
of Christianity, may have been copied only three or four decades after
John’s Gospel was written. The handwriting style in this justifiably famous
fragment is closer to a literary bookhand than to reformed documentary.
And this is also true of ∏64 + ∏67 + ∏4 (from a four-gospel codex), ∏77,
∏90 and ∏104 (Matthew), and ∏66 and ∏90 (John). These are the papyri
with the strongest claims to dating from the latter decades of the second
century, the very time when Irenaeus was underlining the authority of the
gospels for the life of the church and their status as Scripture. No wonder
these papyri were copied carefully, in stylish handwriting.
 Bibliography

P R I MA RY SO U RC E S
(Editions and translations of the main non-canonical writings discussed. For Philo,
Josephus, Cicero, and Martial, I have used the Loeb Classical Library editions.)

the d ea d s ea s c ro l l s
The Dead Sea Scrolls: English and Hebrew, ed. and trans. Florentino Garcı́a Martı́nez
and Eibert J. C. Tigchelaar, Vols. i and ii (Leiden: Brill, 1997–8)

pa p yr i a n d in s c r ip tio n s
The Oxyrhynchus Papyri (London: Egypt Exploration Fund, 1898–[2003])
Brunt, P. A. and Moore, J. M., eds., Res Gestae Divi Augusti (Oxford: Oxford
University Press, 1967)
Hagedorn, U. and D. and Youtie, L. C. and H. C., eds., P. Petaus: Das Archiv des
Petaus, Pap. Colon., Vol. iv (Cologne and Opladen, 1969)
Kenyon, F. G., The Chester Beatty Biblical Papyri: Descriptions and Texts of Twelve
Manuscripts on Papyrus of the Greek Bible, Vols. i and ii (London: Emery
Walker, 1933)
Laffi, U., ‘Le iscrizioni relative all’introduzione nel 9 a.C. del nuovo calendario
della Provincia d’Asia’, Studi Classici e Orientali 16 (1968) 5–98
Sherk, R. K., Roman Documents from the Greek East: Senatus Consulta and Epistula
to the Age of Augustus (Baltimore: Johns Hopkins, 1969)

a p o s to l ic father s
Holmes, M. W., ed., The Apostolic Fathers: Greek Texts and English Translations,
2nd edn (Grand Rapids: Baker, 1992)

j u s tin m a rt yr
Marcovich, M., Iustini Martyris Apologiae Pro Christianis, Patristische Texte und
Studien 38 (Berlin: de Gruyter, 1994)
Iustini Martyris Dialogus cum Tryphone, Patristische Texte und Studien 47
(Berlin: de Gruyter, 1997)

207
208 Bibliography
iren a eu s
Rousseau, A. and Doutreleau, L., eds., Irénée de Lyon. Contre les hérésies, iii, Sources
Chrétiennes 211 (Paris: Cerf, 1974)

m u r ato r ia n f r agm e n t
Lietzmann, H., Das Muratorische Fragment und die monarchianischen Prologue zu
den Evangelien, Kleine Texte, i (Bonn, 1902)
Tregelles, S. P., Canon Muratorianus (Oxford: Clarendon, 1867)

ac ts o f tho m a s
Bonnet, M., ed., Acta Apostolorum Apocrypha ii/2 (Hildesheim, 1903, repr. 1959)
for the Greek text.

o r igen
Chadwick, H., Origen: Contra Celsum (Cambridge: Cambridge University Press,
1953)

S E C O N D A RY L I T E R AT U R E
Abramowski, L., ‘Die “Erinnerungen der Apostel” bei Justin’, in P. Stuhlmacher,
ed., Das Evangelium und die Evangelien (Tübingen: Mohr, 1983), pp. 341–54
Aland, K., Repertorium der griechischen christlichen Papyri, Vol. i: Biblische Papyri
(Berlin and New York: de Gruyter, 1976)
Aland, K., ed., Kurzgefasste Liste der griechischen Handschriften des Neuen Testaments,
2nd edn (Berlin: de Gruyter, 1994)
Albl, M. C., ‘And Scripture Cannot be Broken’: The Form and Function of the Early
Christian Testimonia Collections (Leiden: Brill, 1999)
Alexander, L. A., ‘Ancient Book Production and the Circulation of the Gospels’,
in R. J. Bauckham, ed., The Gospels for All Christians: Rethinking the Gospel
Audiences (Grand Rapids: Eerdmans, 1998), pp. 71–111
‘The Living Voice: Scepticism towards the Written Word in Early Christian and
in Graeco-Roman Texts’, in D. J. A. Clines, S. E. Fowl, and S. E. Porter, eds.,
The Bible in Three Dimensions (Sheffield: JSOT, 1990)
Alexander, P. S., ‘Incantations and Books of Magic’, in E. Schürer, The History of
the Jewish People in the Age of Jesus Christ iii.1, ed. G. Vermes, F. Millar, and
M. Goodman (Edinburgh: T. & T. Clark, 1986), §32.vii, pp. 342–79
Alföldy, G., ‘Subject and Ruler, Subjects and Methods: An Attempt at a Conclu-
sion’, in A. Small, ed., Subject and Ruler: The Cult of the Ruling Power in
Classical Antiquity. Papers presented at a conference held in the University of
Alberta on April 13–15th, 1994, to celebrate the 65th anniversary of Duncan
Fishwick ( Journal of Roman Archaeology Supplementary Series 17 (1996))
 Bibliography 209
Aune, D. E., ‘Magic in Early Christianity’, ANRW ii.23.2, ed. W. Haase (Berlin:
de Gruyter, 1980)
Prophecy in Early Christianity and the Ancient Mediterranean World (Grand
Rapids: Eerdmans, 1983), pp. 153–89
Baarda, T., Essays on the Diatessaron (Kampen: Kok Pharos, 1994)
Babcock, W., ed., Paul and the Legacies of Paul (Dallas: Southern Methodist
University Press, 1990)
Backhouse, J., The Lindisfarne Gospels (London: Phaidon, 1981)
Bammel, E., ‘Der Jude des Celsus’, in his Judaica: Kleine Schriften I (Tübingen:
Mohr, 1986), pp. 265–83
Barclay, J. M. G., Jews in the Mediterranean Diaspora (Edinburgh: T. & T. Clark,
1996)
‘Mirror-Reading a Polemical Letter: Galatians as a Test Case’, JSNT 31 (1987)
73–93
Obeying the Truth (Edinburgh: T. & T. Clark, 1988)
Barton, J., The Spirit and the Letter: Studies in the Biblical Canon (London, SPCK,
1997)
Bauckham, R. J., ed., The Gospels for All Christians: Rethinking the Gospel Audiences
(Grand Rapids: Eerdmans, 1998)
Bellinzoni, A. J., ‘The Gospel of Matthew in the Second Century’, SC 9 (1992)
197–258
The Sayings of Jesus in the Writings of Justin Martyr (Leiden: Brill, 1967)
Benko, S., ‘Pagan Criticism of Christianity during the First Two Centuries’, ANRW
ii.23.2, ed. W. Haase (Berlin: de Gruyter, 1980), pp. 1055–1118
Benoit, A., Saint Irénée. Introduction à l’étude de sa théologie (Paris: Presses Univer-
sitaires de France, 1960)
Betz, H. D., Galatians, Hermeneia Commentary (Philadelphia: Fortress, 1979)
Bingham, D. J., Irenaeus’s Use of Matthew’s Gospel in Adversus Haereses (Leuven:
Peeters, 1998)
Birley, R., Roman Records from Vindolanda on Hadrian’s Wall, 3rd edn, Roman
Army Museum Publications for the Vindolanda Trust (Carvoran: Greenhead,
1999)
Blanchard, Alain, ed., Les Débuts du codex, Bibliologia 9 (Brepols: Turnhout, 1989)
Blanchard, Y.-M., Aux sources du canon, le témoinage d’Irénée, Cogitatio Fidei 174
(Paris: Cerf, 1993)
Blanck, Horst, Das Buch in der Antike (Munich: C. H. Beck, 1992)
Borg, M. et al., eds., The Lost Gospel Q: The Original Sayings of Jesus (Berkeley,
Calif.: Ulysses, 1996)
Bovon, F., ‘La Structure canonique de l’Evangile et de l’Apôtre’, Cristianesimo nella
Storia 15 (1994) 559–76
Bowman, A. K., Life and Letters on the Roman Frontier: Vindolanda and its People
(London: British Museum, 1994)
Bowman, A. K. and Thomas, J. D., Vindolanda: The Latin Writing-Tablets, Bri-
tannia Monograph Series 4 (London: Society for the Promotion of Roman
Studies, 1983)
210 Bibliography
The Vindolanda Writing-Tablets: Tabulae Vindolandansenses ii (London: British
Museum, 1994)
Brent, A., The Imperial Cult and the Development of Church Order (Leiden: Brill,
1999)
Breytenbach, C., Paulus und Barnabas in der Provinz Galatien. Studien zu Apos-
telgeschichte 13f.; 16, 6; 18, 23 und den Adressaten des Galaterbriefes (Leiden:
Brill, 1996)
Brooke, G., Exegesis at Qumran: 4QFlorilegium in its Jewish Context (Sheffield:
JSOT, 1985)
Brown, R. E., The Epistles of John, Anchor Bible (Garden City: Doubleday, 1982)
Burridge, R. A., Four Gospels, One Jesus? (London: SPCK, 1994)
What are the Gospels? A Comparison with Graeco-Roman Biography (Cambridge:
Cambridge University Press, 1992; 2nd edn Grand Rapids: Eerdmans, 2004)
Campenhausen, H. von, Die Entstehung der christlichen Bibel (Tübingen: Mohr,
1968); E. tr. The Formation of the Christian Bible (Philadelphia: Fortress;
London: Black, 1972)
Carleton Paget, James, ‘Some Observations on Josephus and Christianity’, JTS 52
(2001) 539–624
Carroll, K. L., ‘The Creation of the Fourfold Gospel’, BJRL 37 (1954–5) 68–
77
Cavallo, G., and Maehler, H., Greek Bookhands of the Early Byzantine Period,
ad 300–88, Bulletin Supplement 47 (London: University of London Institute
of Classical Studies, 1987)
Chilton, B. D. ‘Jésus, le mamzer (Mt 1.18)’, NTS 46 (2000) 222–7
Collins, J. J., ‘Jesus, Messianism and the Dead Sea Scrolls’, in J. H. Charlesworth,
H. Lichtenberger, and G. S. Oegema, eds., Qumran–Messianism: Studies on
the Messianic Expectations in the Dead Sea Scrolls (Tübingen: Mohr–Siebeck,
1998), pp. 100–19
The Scepter and the Star (New York: Doubleday, 1995)
Cosgrove, C. H., ‘Justin Martyr and the Emerging Christian Canon: Observations
on the Purpose and Destination of the Dialogue with Trypho’, Vig. Chr. 36
(1982) 209–32
Coulmas, Florian, ed., The Handbook of Sociolinguistics (Oxford: Blackwell, 1997)
Crossan, J. D., The Historical Jesus: The Life of a Mediterranean Jewish Peasant
(Edinburgh: T. & T. Clark, 1991)
Croy, N. Clayton, The Mutilation of Mark’s Gospel (Nashville: Abingdon, 2003)
Cullmann, O., ‘Die Pluralität der Evangelien als theologisches Problem im Alter-
tum’, Theologische Zeitschrift 1 (1945), 23–42; E. tr. in Cullmann’s The Early
Church, ed. A. J. B. Higgins (London: SCM, 1956)
Danker, F. W., Benefactor (St Louis, Miss.: Clayton, 1982)
Davies, W. D., The Setting of the Sermon on the Mount (Cambridge: Cambridge
University Press, 1963)
Deissmann, A., Light from the Ancient East (London: Hodder and Stoughton, E.
tr. 1910)
 Bibliography 211
Denney, J., Jesus and the Gospels: Christianity Justified in the Mind of Christ
(London: Hodder & Stoughton, 1908)
Dibelius, M., From Tradition to Gospels (London: Ivor Nicholson and Watson, E.
tr. 1934)
Dodd, C. H., More New Testament Studies (Manchester: Manchester University
Press, 1968)
Dohmen, C. and Oeming, M., Biblischer Kanon, warum und wozu?: eine Kanon-
theologie (Freiburg: Herder, 1992)
Donfried, K. P., Paul, Thessalonica and Early Christianity (London: T. & T. Clark,
2002)
Ehrhardt, A. T., The Framework of the New Testament Stories (Manchester: Manch-
ester University Press, 1964)
Ehrman, B. D. and Holmes, M. W., The Text of the New Testament in Contemporary
Research (Grand Rapids: Eerdmans, 1995)
Elliott, J. K., The Apocryphal New Testament (Oxford: Clarendon Press, 1993)
A Bibliography of Greek New Testament Manuscripts (Cambridge: Cambridge
University Press, 1989; 2nd edn, 2000)
‘Five New Papyri of the New Testament’, NovT 41 (1999) 209–13
‘Manuscripts, the Codex and the Canon’, JSNT 63 (1996) 105–23
‘Seven Recently Published New Testament Fragments from Oxyrhynchus’,
NovT 42 (2000) 209–13
Elliott, W. J. and Parker, D. C., eds., The New Testament in Greek, Vol. iv, The
Gospel According to St John: The Papyri (Leiden: Brill, 1995)
Epp, E. J., ‘The Codex and Literacy in Early Christianity and at Oxyrhynchus:
Issues Raised by Harry Y. Gamble’s Books and Readers in the Early Church’,
Critical Review of Books in Religion 10 (1997) 15–37
‘New Testament Papyrus Manuscripts and Letter Carrying in Greco-Roman
Times’, in B. A. Pearson, ed., The Future of Early Christianity, FS H. Koester
(Minneapolis: Fortress, 1991), pp. 35–56
‘The Papyrus Manuscripts of the New Testament’, in B. D. Ehrman and
M. W. Holmes, eds., The Text of the New Testament in Contemporary Research,
FS B. M. Metzger (Grand Rapids: Eerdmans, 1995), pp. 3–21
Evans, C. A., ‘Jesus and the Dead Sea Scrolls’, in P. W. Flint and J. C. Vanderkam,
eds., The Dead Sea Scrolls after Fifty Years (Leiden: Brill, 1999)
Falcetta, A., ‘A Testimony Collection in Manchester: Papyrus Rylands Greek 460’
BJRL 83 (2002) 3–19
Farmer, W. R. and Farkasfalvy, D. M., The Formation of the New Testament Canon
(New York, 1983)
Ferguson, E., ‘Canon Muratori: Date and Provenance’, Studia Patristica 18 (1982)
677–83
Fitzmyer, J. A., Essays on the Semitic Background of the New Testament (London:
Chapman, 1971)
Ford, D. F. and Stanton, G. N., Scripture and Theology: Reading Texts, Seeking
Wisdom (London: SCM, 2003)
212 Bibliography
Frankemölle, H., Evangelium. Begriff und Gattung, 2nd edn (Stuttgart: Katholisches
Bibelwerk, 1994)
Jahwebund und Kirche Christi (Münster: Aschendorff, 1974)
French, D., ‘Acts and the Roman Roads of Asia Minor’, in D. W. J. Gill and C.
Gempf, eds., The Book of Acts in its Graeco-Roman Setting (Grand Rapids:
Eerdmans, 1994), pp. 49–58
Frickenschmidt, D., Evangelium als Biographie. Die vier Evangelien im Rahmen
antiker Erzählungskunst (Tübingen and Basle: A. Francke, 1997)
Friedrich G., art. eÉagg”lion in Theological Dictionary of the New Testament,
ed. G. Kittel, Vol. ii (Grand Rapids: Eerdmans, E. tr. 1964), pp. 707–
37
Friesen, S. J., Twice Neokoros. Ephesus, Asia and the Cult of the Flavian Imperial
Family (Leiden: Brill, 1993)
Funk, R. W., ed., The Gospel of Jesus according to the Jesus Seminar (Sonoma, Calif.:
Polebridge, 1999)
Gamble, H. Y., Books and Readers in the Early Church: A History of Early Christian
Texts (New Haven and London: Yale University Press, 1995)
Gerhardsson, B., Memory and Manuscript: Oral Tradition and Written Transmis-
sion in Rabbinic Judaism and Early Christianity, 2nd edn (Grand Rapids:
Eerdmans, 1998)
Goodspeed, E. J., The Formation of the New Testament (Chicago: University of
Chicago Press, 1937)
Gradel, I., Emperor Worship and Roman Religion (Oxford: Clarendon, 2002)
Grant, R. M., The Earliest Lives of Jesus (London: SPCK, 1961)
Guelich, R., Mark, Word Biblical Commentary (Dallas: Word, 1989)
Gundry, R., ‘No NU in Line 2 of 7Q5: A Final Disidentification of 7Q5 with Mark
6:52-3’, JBL (1999) 698–702
Haacker, K., ‘Das Bekenntnis des Paulus zur Hoffnung Israels nach der Apos-
telgeschichte des Lukas’, NTS 31 (1985) 437–51
Haelst, J. van, Catalogue des papyrus littéraires juifs et chrétiens, Papyrologie i (Paris:
Publications de la Sorbonne, 1976)
‘Les Origines du codex’, in A. Blanchard, ed., Les Débuts du codex, Bibliologia
9 (Brepols: Turnhout, 1989), pp. 12–35
Hahneman, G. M., The Muratorian Fragment and the Development of the Canon
(Oxford: Clarendon, 1992)
Halverson, J., ‘Oral and Written Gospel: A Critique of Werner Kelber’, NTS 40
(1994) 180–95
Haran, M., ‘Codex, Pinax and Writing Slat’, Scripta Classica Israelica 15 (1996)
212–22
Harnack, A., Judentum und Judenchristentum in Justins Dialog mit Tryphon (Leipzig,
1930)
Marcion: Das Evangelium vom fremden Gott (Leipzig: J. C. Hinrichs, 1921)
The Origin of the New Testament and the Most Important Consequences of the New
Creation (London and New York, 1925)
 Bibliography 213
Harris, W. F., ‘Why did the Codex Supplant the Book-Roll?’, in J. Monfasani and
R. G. Musto, eds., Renaissance Society and Culture, FS E. F. Price Jr. (New
York: Italica Press, 1991), pp. 71–85
Harrison, J. R., ‘Paul and the Imperial Gospel at Thessaloniki’, JSNT 25 (2002)
71–96
‘Paul, Eschatology and the Augustan Age of Grace’, TB 50 (1999) 79–91
‘Paul’s Language of Grace (C†riv) in its Graeco-Roman Context’, Macquarie
University Ph.D. 1996, to be published in the WUNT series by J. C. B. Mohr,
Tübingen
Hays, R. B., ‘Christology and Ethics in Galatians: The Law of Christ’, CBQ 49
(1987) 268–90
Head, P. M., ‘Some Recently Published NT Papyri from Oxyrhynchus: An
Overview and Preliminary Assessment’, TB 51 (2000) 1–16
Heckel, T. K., Vom Evangelium des Markus zum viergestaltigen Evangelium
(Tübingen: Mohr, 1999)
Hengel, M., The Four Gospels and the One Gospel of Jesus Christ (London: SCM,
2000)
‘The Titles of the Gospels and the Gospel of Mark’, in Studies in the Gospel of
Mark (London: SCM, 1985)
Hengel, M. and Schwemer, A. M., Paul between Damascus and Antioch (London:
SCM, 1997)
Henne, P., ‘La Datation du Canon de Muratori ’, RB 100 (1993) 54–75
Hill, C. E., ‘Justin and the New Testament Writings’, in E. A. Livingstone, ed.,
Studia Patristica 30 (Leuven: Peeters, 1997)
Horbury, W., ‘The Wisdom of Solomon in the Muratorian Fragment’, JTS 45
(1994) 149–59
Horsley, G. H. R., ‘The “Good News” of a Wedding’, New Documents Illustrat-
ing Early Christianity, Vol. iii (Sydney: The Ancient History Documentary
Research Centre, Macquarie University, 1983), pp. 10–15
Horsley, R. A., Paul and Politics: Ekklesia, Imperium, Interpretation (Harrisburg,
Pa.: Trinity Press International, 2000)
Hultgren, S., Narrative Elements in the Double Tradition (Berlin: de Gruyter, 2002)
Hurtado, L. W., ‘The Origin of the Nomina Sacra: A Proposal’, JBL 117 (1998)
655–73
Hyldahl, N., ‘Hesesipps Hypomnemata’, ST 14 (1960) 70–113
Kaestli, J.-D., ‘La Place du Fragment de Muratori dans l’histoire du canon. A propos
de la thèse de Sundberg et Hahneman’, Cristianesimo nella Storia 15 (1994)
609–34
Kingsbury, J. D., Matthew: Structure, Christology, Kingdom (London: SPCK, 1975)
Knoch, O., ‘Kenntnis und Verwendung des Matthäus-Evangeliums bei den Apos-
tolischen Vätern’, in L. Schenke, ed., Studien zum Matthäusevangeliums, FS
W. Pesch (Stuttgart: Katholisches Bibelwerk, 1988)
Knox J., Marcion and the New Testament (Chicago: University of Chicago Press,
1942)
214 Bibliography
Koch, D.-A., Die Schrift als Zeuge des Evangeliums (Tübingen: Mohr, 1986)
Koester, H., Ancient Christian Gospels: Their History and Development (London:
SCM and Philadelphia: Trinity, 1990)
‘From the Kerygma-Gospel to Written Gospels’, NTS 35 (1989) 361–81
History and Literature of Early Christianity (Philadelphia: Fortress, 1982)
‘The Text of the Synoptic Gospels in the Second Century’, in W. L. Petersen, ed.,
Gospel Traditions in the Second Century (Notre Dame and London: University
of Notre Dame Press, 1989), pp. 28–33
Köhler, W., Die Rezeption des Matthäusevangeliums in der Zeit vor Irenäus, WUNT
24 (Tübingen: Mohr, 1987)
Lane Fox, R., ‘Literacy and Power in Early Christianity’, in A. K. Bowman and G.
Woolf, eds., Literacy and Power in the Ancient World (Cambridge: Cambridge
University Press, 1994), pp. 126–48
Lapide, Pinchas, The Resurrection of Jesus (London: SPCK, 1984)
Layton, B., ed., Nag Hammadi Codex II, 2–7 (Leiden: Brill, 1989)
Levick, B., Roman Colonies in Southern Asia Minor (Oxford: Oxford University
Press, 1967)
Lieberman, S., Hellenism in Jewish Palestine, 2nd edn (New York: Jewish Theolo-
gical Seminary, 1962)
Lim, T. H., Holy Scripture in the Qumran Communities and the Pauline Letters
(Oxford: Clarendon, 1997)
Llewelyn, S. R., New Documents Illustrating Early Christianity, Vol. vii (Sydney:
The Ancient History Documentary Research Centre, Macquarie University,
1994)
Longenecker, B. W., ed., Narrative Dynamics in Paul (Louisville: Westminster John
Knox, 2002)
Luz, U., Das Evangelium nach Matthäus, EKK i/1, 5th edn (Düsseldorf and Zurich:
Benziger, 2002); Matthew (Edinburgh: T. & T. Clark, E. tr. of 1st edn, 1989)
McCormick, M., ‘The Birth of the Codex and the Apostolic Life-Style’, Scriptorium
39 (1985) 150–8
‘Typology, Codicology and Papyrology’, Scriptorium 35 (1981) 331–4
McDonald, J. I. H., ‘Questioning and Discernment in Gospel Discourse: Com-
municative Strategy in Matthew 11.2-9’, in B. Chilton and C. A. Evans, eds.,
Authenticating the Words of Jesus (Leiden: Brill, 1999)
McKnight, S., ‘Calling Jesus Mamzer’, Journal for the Study of the Historical Jesus 1
(2003) 73–103
Maier, J., Jesus von Nazareth in der talmudischen Überlieferung (Darmstadt:
Wissenschaftliche Buchgesellschaft, 1978)
Malina, Bruce J. and Neyrey, Jerome H., Calling Jesus Names: The Social Value of
Labels in Matthew (Sonoma, Calif.: Polebridge, 1988)
Marcus, J., ‘Mark – Interpreter of Paul’, NTS 46 (2000) 473–87
Mark 1–8, Anchor Bible 27 (New York: Doubleday, 2000)
Marshall, I. H., The Pastoral Epistles, ICC (Edinburgh, T. & T. Clark, 1999)
Martin, V., Papyrus Bodmer ii, Evangile de Jean, chs. 1–14 (Cologne and Geneva:
Bibliotheca Bodmeriana, 1956), pp. 14–17
 Bibliography 215
Marxsen, W., Mark the Evangelist (E. tr. Nashville: Abingdon, 1969)
Meeks, W., The First Urban Christians: The Social World of the Apostle Paul (New
Haven: Yale, 1983)
Meggitt, J., ‘Taking the Emperor’s Clothes Seriously: The New Testament and the
Roman Emperor’, in Christine E. Joynes, ed., The Quest for Wisdom: Essays
in Honour of Philip Budd (Cambridge: Orchard Academic, 2002)
Meier, J. P., ‘Jesus in Josephus: A Modest Proposal’, CBQ 52 (1990) 76–103
Merkel, H., Die Pluralität der Evangelien als theologisches und exegetisches Problem
in der alten Kirche (Bern: Peter Lang, 1978)
Die Widersprüche zwischen den Evangelien. Ihre polemische und apologetische Be-
handlung in der alten Kirche bis zu Augustin (Tübingen: Mohr, 1971)
Metzger, B. M., The Canon of the New Testament: Its Origin, Development, and
Significance (Oxford: Clarendon, 1987)
Millar, F., ‘The Impact of Monarchy’, in F. Millar and E. Segal, eds., Caesar
Augustus: Seven Aspects (Oxford: Clarendon, 1984)
Millard, A. R., Reading and Writing in the Time of Jesus (Sheffield: Sheffield
Academic Press, 2000)
Mitchell, M. M., ‘Rhetorical Shorthand in Pauline Argumentation: The Functions
of “the Gospel” in the Corinthian Correspondence’, in L. A. Jervis and P.
Richardson, eds., Gospel in Paul, FS R. N. Longenecker (Sheffield: JSOT Press,
1994), pp. 63–88
Mitchell, S., Anatolia: Land, Men and Gods in Asia Minor, Vols. i and ii (Oxford:
Oxford University Press, 1993)
Mitchell, S. and Waelkens, M., eds., Pisidian Antioch: The Site and its Monuments
(London, Duckworth, 1998)
Morgan, R. C., ‘The Hermeneutical Significance of Four Gospels’, Interpretation
33 (1979) 376–88
Muddiman, J., The Epistle to the Ephesians, Black’s NT Commentaries (London
and New York: Continuum, 2001)
Oakes, Peter, Philippians: From People to Letter (Cambridge: Cambridge University
Press, 2001)
Osborn, E., Justin Martyr (Tübingen: Mohr, 1973)
Parsons, M. C. and Pervo, R. I., Rethinking the Unity of Luke and Acts (Minneapolis:
Augsburg Fortress, 1993)
Petersen, W. L., Gospel Traditions in the Second Century (Notre Dame and London:
University of Notre Dame Press, 1989)
Tatian’s Diatessaron: Its Creation, Dissemination, Significance, and History in
Scholarship (Leiden: Brill, 1994)
Pickering, S. R., Recently Published New Testament Papyri: P89–P95, Papyrology and
Historical Perspectives 2 (Sydney: The Ancient History Documentary Research
Centre, Macquarie University, 1991)
Pilhofer, P., ‘Luke’s Knowledge of Antioch’, in T. Drew Bear, M. Tashalan, and
C. M. Thomas, eds., Actes du Ier Congrès International sur Antioche de Pisi-
die (Lyons and Paris: Université Lumière-Lyon 2 and Diffusion de Boccard,
2000)
216 Bibliography
Price, S. R. F., Rituals and Power: The Roman Imperial Cult in Asia Minor
(Cambridge: Cambridge University Press, 1984)
Pryor, J. W., ‘Justin Martyr and the Fourth Gospel’, The Second Century 9 (1992)
153–69
Reeve, Anne, Erasmus’s Annotations on the New Testament. Galatians to the
Apocalypse: Facsimile of the Final Latin Text with All Earlier Variants (Leiden:
Brill, 1993)
Rendel Harris, J., Testimonies (2 vols., Cambridge: Cambridge University Press,
1916–20)
Richards, E. R., The Secretary in the Letters of Paul (Tübingen: Mohr, 1991)
Riesner, R., Paul’s Early Period: Chronology, Mission Strategy, Theology (Grand
Rapids: Eerdmans, 1998)
Robbins, G. A., ‘Muratorian Fragment’, in D. N. Freedman, ed., The Anchor Bible
Dictionary (New York: Doubleday 1992)
Roberts, C. H., ‘Books in the Graeco-Roman World’, in The Cambridge History
of the Bible, ed. P. R. Ackroyd and C. F. Evans (Cambridge: Cambridge
University Press, 1970), pp. 12–23
‘The Codex’, Proceedings of the British Academy 40 (1954) 169–204
Manuscript, Society and Belief in Early Christian Egypt (London: Oxford Uni-
versity Press for the British Academy, 1979)
An Unpublished Fragment of the Fourth Gospel in the John Rylands Library
(Manchester: Manchester University Press, 1935)
Roberts, C. H. and Skeat, T. C., The Birth of the Codex (London: Oxford University
Press for the British Academy, 1983)
Sanders, J. N., The Fourth Gospel in the Early Church (Cambridge: Cambridge
University Press, 1943)
Sato, M., Q und Prophetie. Studien zur Gattungs- und Traditionsgeschichte der Quelle
Q, WUNT 2. Reihe, Vol. 29 (Tübingen: Mohr, 1988)
Schenk, W., Die Sprache des Matthäus (Göttingen:Vandenhoeck & Ruprecht, 1987)
Schnackenburg, R., Jesus in the Gospels: A Biblical Christology (Louisville: West-
minster John Knox, E. tr. 1995)
Schneemelcher, W., ed., New Testament Apocrypha, Vol. ii (Cambridge: Clarke,
1992)
Schniewind, J., Euangelion. Ursprung und erste Gestalt des Begriffs Evangelium,
Vols. i–ii (Gütersloh, 1927–31)
Schoedel, W. R., Ignatius of Antioch, Hermeneia (Philadelphia: Fortress, 1985)
Shum, S.-L., Paul’s Use of Isaiah in Romans (Tübingen: Mohr Siebeck, 2002)
Sirat, C., ‘Le Livre hébreu dans les premiers siècles de nôtre ère: le témoinage des
textes’, in A. Blanchard, ed., Les Débuts du codex (Brepols: Turnhout, 1989)
Skarsaune, O., The Proof from Prophecy. A Study in Justin Martyr’s Proof-Text
Tradition: Text-Type, Provenance, Theological Profile (Leiden: Brill, 1987)
Skeat, T. C., ‘The Codex Sinaiticus, the Codex Vaticanus, and Constantine’, JTS
50 (2000) 583–625
‘A Codicological Analysis of the Chester Beatty Papyrus Codex of Gospels and
Acts (P45)’, Hermathena 155 (1993) 27–43
 Bibliography 217
‘Irenaeus and the Four-Gospel Canon’, NovT 34 (1992) 193–9
‘The Oldest Manuscript of the Four Gospels?’, NTS 43 (1997) 1–34
‘The Origin of the Christian Codex’, ZPE 102 (1994) 263–8
Small, A., ed., Subject and Ruler: The Cult of the Ruling Power in Classical Antiquity
( Journal of Roman Archaeology Supplementary Series 17 (1996))
Smith, M., Jesus the Magician (London: Victor Gollancz, 1978)
Spallek, A. J., ‘The Origin and Meaning of EÉagg”lion in the Pauline Corpus’,
CTQ 57 (1993) 177–90
Stanley, C. D., Paul and the Language of Scripture (Cambridge: Cambridge
University Press, 1992)
Stanton, G. N., ‘Form Criticism Revisited’, in M. Hooker and C. J. A. Hickling,
eds., What about the New Testament? FS C. F. Evans (London: SCM, 1975),
pp. 13–27
A Gospel for a New People: Studies in Matthew (Edinburgh: T. & T. Clark, 1992)
Gospel Truth? New Light on Jesus and the Gospels (London: HarperCollins, 1995)
The Gospels and Jesus, 2nd edn (Oxford: Oxford University Press, 2002)
Jesus of Nazareth in New Testament Preaching (Cambridge: Cambridge University
Press, 1974).
‘Matthew: b©blov, eÉagg”lion, or b©ov?’, in F. Van Segbroeck et al., eds., The
Four Gospels 1992, FS Franz Neirynck, Vol. ii (Leuven: Leuven University Press,
1992), pp. 1187–1202
‘On the Christology of Q’, in B. Lindars and S. S. Smalley, eds., Christ and
Spirit in the New Testament (Cambridge: Cambridge University Press, 1973),
pp. 27–42
‘Revisiting Matthew’s Communities’, in SBL Seminar Papers 1994, ed. E. H.
Lovering Jr. (Atlanta: Scholars Press, 1994), pp. 9–23
‘Stephen in Lucan Perspective’, in E. A. Livingstone, ed., Studia Biblica III
(Sheffield: JSOT Press, 1980), pp. 345–60
‘The Two Parousias of Christ: Justin Martyr and Matthew’, in M. C. de
Boer, ed., From Jesus to John, FS M. de Jonge (Sheffield: JSOT Press, 1993),
pp. 183–96
Strange, W. A., The Problem of the Text of Acts (Cambridge: Cambridge University
Press, 1992)
Strecker, G., ‘Das Evangelium Jesu Christi’, in G. Strecker, ed., Jesus Christus in
Historie und Theologie, FS H. Conzelmann (Tübingen: Mohr, 1975), pp. 503–
48
Stuhlmacher, P., Das paulinische Evangelium (Göttingen: Vandenhoeck &
Ruprecht, 1968)
Stuhlmacher, P., ed., Das Evangelium und die Evangelien (Tübingen: Mohr, 1983)
The Gospel and the Gospels (Grand Rapids: Eerdmans, 1991)
Sundberg, A. C., ‘Canon Muratori: A Fourth Century List’, HTR 66 (1973) 1–41
‘Towards a Revised History of the New Testament Canon’, Studia Evangelica
4/1 (1968), pp. 452–61
Thiede, C. and d’Ancona, M., The Jesus Papyrus (London: Weidenfeld & Nicolson,
1996)
218 Bibliography
Thompson, M. B., Clothed with Christ: The Example and Teaching of Jesus in Romans
12.1-15.3 (Sheffield: Sheffield Academic Press, 1991)
Trebilco, Paul, Jewish Communities in Asia Minor (Cambridge: Cambridge Uni-
versity Press, 1991)
Trobisch, D., Die Endredaktion des Neuen Testaments. Eine Untersuchung zur Entste-
hung der christlichen Bibel (Freiburg: Universitätsverlag; Göttingen: Vanden-
hoeck & Ruprecht, 1996)
Paul’s Letter Collection: Tracing the Origins (Minneapolis: Fortress, 1994)
Tuckett, C. M., ‘“Nomina Sacra”: Yes and No?’, in J.-M. Auwers and H. J.
de Jonge, eds., The Biblical Canons (Leuven: Peeters, 2003), pp. 431–
58
‘Scripture and Q’, in C. M. Tuckett, ed., The Scriptures in the Gospels (Leuven:
University Press, 1997), pp. 20–6
Turner, E. G., Greek Manuscripts of the Ancient World (Oxford: Oxford University
Press, 1971; 2nd edn, 1987)
Greek Papyri, 2nd edn (Oxford: Clarendon, 1980)
The Typology of the Early Codex (Philadelphia: University of Pennsylvania Press,
1977)
Unnik, W. C. van, ‘De la règle Mžte prosqe±nai mžte ˆjele±n’, Vig. Chr. 3 (1949)
1–36
Verheyden, J., ‘The Canon Muratori: A Matter of Dispute’, in J.-M. Auwers and
H. J. de Jonge, eds., The Biblical Canons (Leuven: Peeters, 2003), pp. 487–
586
Vocke, H., ‘Papyrus Magdalen 17 – Weitere Argumente gegen die Frühdatierung
des angeblichen Jesus-Papyrus’, ZPE 113 (1996) 153–7
Wachtel, K., ‘P64/67: Fragmente des Matthäusevangeliums aus dem 1.
Jahrhundert?’, ZPE 107 (1995) 73–80
Wengst, K., Schriften des Urchristentums, Vol. ii (Darmstadt: Wissenschaftliche
Buchgesellschaft, 1984)
Whittaker, Molly, Jews and Christians: Graeco-Roman Views (Cambridge:
Cambridge University Press, 1984)
Wilken, R. L., The Christians as the Romans Saw Them (New Haven and London:
Yale University Press, 1984)
Williams, A. Lukyn, Justin Martyr: The Dialogue with Trypho (London: SPCK,
1930)
Winger, M., ‘Act One: Paul Arrives in Galatia’, NTS 48 (2002) 548–67
‘The Law of Christ’, NTS 46 (2000) 537–46
Winter, B. W., ‘The Imperial Cult and Early Christians in Roman Galatia (Acts
XIII 13-50 and Galatians VI 11-18’, in T. Drew-Bear, M. Tashalan, and
C. M. Thomas, eds., Actes du Ier Congrès International sur Antioche de Pisidie
(Université Lumière-Lyon 2 and Diffusion de Boccard, 2002)
Witulski, T., Die Adressaten des Galaterbriefes. Untersuchungen zur Gemeinde von
Antiochia ad Pisidiam (Göttingen: Vandenhoeck & Ruprecht, 2000)
Wright, N. T., ‘Five Gospels but No Gospel’, in B. Chilton and C. A. Evans, eds.,
Authenticating the Activities of Jesus (Leiden: Brill, 1999)
 Bibliography 219
Yadin, Y., ‘Expedition D’, Israel Exploration Journal 11 (1961) 41–2
Zahn, Th., Geschichte des neutestamentlichen Kanons (Erlangen: Deichert, 1888)
Grundriss der Geschichte des neutestamentlichen Kanons, 2nd edn (Leipzig:
Deichert, 1904)
Zanker, P., The Power of Images in the Age of Augustus (Ann Arbor: University of
Michigan Press, 1990)
 Index of passages cited

Old Testament

Genesis
2.10–14 91

Leviticus
19.18 115, 122

Deuteronomy
13 130, 131, 136, 139, 140
13.5 129
13.8 130
18 130, 131
18.9–14 131
18.15–22 131
18.15 141

Judges
12.5–6 51

II Samuel
4.10 13, 20
7.12 46
18.20, 25, 27 13
18.22 13

I Kings
17.21–2 151

II Kings
4.34–5 151
7.9 13

Psalms
2.7 46
22 76
40.10 13
68.11 13
96.2–3 13

220
 Index of references 221
Isaiah
1.3 97
1.16–20 97
3.9–11 98
5.18–20 98
9.1 116
29.18 15
35.1–7 129
35.5–6 15
40.9 13
52.5 98
52.7 13, 17, 21
58.6 184
60.6 13
61 14–15, 17–18, 99
61.1 184
61.1–2 11, 13, 15, 16, 17, 20, 22

Jeremiah
7.11 98

Ezekiel
1 65

Micah
5.2 98

Old Testament apocrypha

Sirach
29.1 116

Testament of Levi
16.3 135

Tobit
14.9 116

New Testament

Matthew
1.23 116
2.5 98
4.16 116
4.17 116
4.22 116
4.23 57, 58, 60
5.17–19 116
5.21–48 116
6.9–11 55
7.12 116
7.15 105
222 Index of references
Matthew (cont.)
7.21 116
9.34 142
9.35 57, 58, 60
10.25 142
11.2–6 15
11.6 118
11.7–19 138
11.18 146
11.27 67, 76, 100, 108
12.18–21 67–70
12.18 142
12.24 144
12.24, 27 142
12.28 142, 145
12.31–2 142
12.43–5 142
13.18 106
13.19 57, 60
18.15–16 120
19.17 116
21.13 98
22.34–40 115
22.34–9 116
23.4 120
24.14 18, 57, 58, 60, 61
26.13 18, 57, 58, 60
27.62–6 142, 155
27.63–4 142
28.11–15 155
28.13 154
28.13–14 153
28.15 154
28.18–20 115, 143

Mark
1.1–3 190
1.1 18, 56
1.14 18, 56
1.15 18, 19–20, 56
3.17 101
3.19b–35 143, 145
3.22 142, 143, 144, 145
3.28–9 145
4.13–20 60
6.4 146
8.35 18, 19, 20, 56
10.29 18, 19, 20, 56
12.13–17 96
13.9 19
13.10 18, 19, 20, 56, 57
14.9 18, 19, 20, 56, 57
15.44–5 160
16.9–20 190
 Index of references 223
Luke
4.16–21 14
4.18–19 182, 184
4.34 14
4.41 14
4.43 60
5.1 60
6.46 116
7.19, 22–3 15, 17–18
7.22 118
7.24–35 138
7.33 146
10.18 67–68, 71
10.22 100, 108
11.15 144
11.19 142
11.20 145
13.33 146
23.2 141
23.5 141
23.14 141
24.19 142
24.21 142
24.25–6 104
24.28 142
24.44–6 104

John
1.19–20 76
1.40 67–70
2.22 61
3.3–5 76, 102
3.3 102
4.19 140
6.14 141
7.12 140, 141
7.25–7 140
7.31 141
7.40 140
7.45 140
7.46 140
7.47 140, 141
7.52 140
8.44 141
8.48 141
8.52 141
9.16–17 141
9.17 140
10.19–21 141
10.20 141
10.21 141
13.34–5 121
14.18–22 159
14.21 160
14.22 159
224 Index of references
Acts
1.8–9 104
4.1–4 157
4.2 158
5.17 157
9.20 46
10.34–43 45
10.36 60
10.40–2 159
11.19 23
11.20 23
11.22–6 23
13.4–14 37
13.6–12 135, 138
13.14–41 45, 46
13.26 45
13.32 46
13.38–9 46
13.47 45
13.50 44
14.6 38
14.8 38
14.12 44
14.15–17 45
14.20–6 38
15.5 116
17.7 47
17.18 156
17.22–31 45
17.31 155
17.32 156
19.11–20 142
20.15–17 32
21.28 113
21.38 136
23.1–10 157
23.6 157, 159
23.9 158
24.15 159
25.19 157
26.2–23 158
26.6 159
26.8 157, 158
26.22–3 157
26.23 158, 159
26.24–6 157
28.20 159

Romans
3.27 115
3.31 115
8.2 115
10.15–16 21
12–14 120
13.8–10 114
15.1–3 113
 Index of references 225
I Corinthians
1.5–6 47
8.5 41
9.18 21
9.20 114
9.21 114, 121
12.2 41
15.1–5 34
15.1 21
15.12 160

II Corinthians
11.7 21
11.32 22

Galatians
1.4 40, 113
1.6–9 36
1.6–8 21
1.7–8 36
1.7 35
1.8–9 22
1.11 21, 40
1.12 40
1.13 44
1.14–17 22
1.15–16 40
1.21–4 23
1.23 1, 46, 56–58
2.4–6 44
2.5 10
2.14 10
2.15–20 40, 113
2.16 46, 112, 113
2.20 40, 113
3.1–5 41
3.5 12, 51
3.13 113
3.21 113
4.1–7 40
4.4–5 34, 113
4.6 41
4.8–20 43
4.8–10 41–43
4.13 22, 36
4.21b 112, 113
5.13–6.10 110
5.14 112, 113, 121, 122
5.16–25 113
5.16 113
5.19–23 114
5.22 121
6.1–5 120
6.2 4, 110–115, 119, 120, 121–123
226 Index of references
Galatians (cont.)
6.11 176
6.12–13 43–45

Philippians
1.13 49
1.14 49
1.27 49
2.6–11 48
3.20 25, 45–46, 48

Colossians
2.16 41

I Thessalonians
1.5 22
1.6 47
1.8 47
2.2 48
2.8–9 48
2.13 47, 48
3.2 48
5.3 47
5.8–9 47

II Thessalonians
3.1 47

I Timothy
3.15 65

II Timothy
2.11 178
3.14–17 178
4.13 177, 178, 181, 182

James
1.25 115
2.8 98
2.12 115

I John
1.5 12, 51
2.3 116
2.6 116
2.10 116
4.1 139

II John
6–10 117

Revelation
2.20–5 139
13.11 139
 Index of references 227
13.13 139
13.14 139
16.13–14 139
16.13 139
19.20 139
20.10 139

Early Jewish writings

Josephus
Antiquitates Judaicae (Ant.)
ii.261–3 136
xviii.3 133–134
xviii.116–19 146
xviii.216–309 24
xx.97 136
xx.169–72 136

De Bello Judaico (Bell.)

ii.420 29, 33
iv.618 29
iv.656 29

Nicolaus of Damascus
90 f 125 (Jacoby) 26

Philo
De Legatione ad Gaium
18 28
22 24, 26
99 24
146–50 26
184–382 24
231 24
231–2 29
346 24
356 29

De Specialibus Legibus
i.315 136

De Vita Mosis
i.277 136

Qumran writings
4Q174 (4Q Testimonia) 160
4Q175 183, 184
4Q176 (4Q Tanhumim) 160, 183
4Q421 15–16
4Q521 20, 61
11Q13 16–17, 21
228 Index of references
Rabbinic writings
m. Kelim 24.7 187–188
m. Sanhedrin 7.10–11 131
b. Sanhedrin 43a 130–131, 140
b. Sanhedrin 107b 131–132

Early Christian writings

Acts of Thomas
48 132
96 132
102 132
106–7 132

II Clement
8.5 54, 79

Didache
8.2 55, 79
11.3 55, 79
15.3–4 55, 79

Epiphanius
Panarion 48.1–13 138

Epistle of Barnabas
2.6 119
10.2 119
10.9 119

Eusebius
Historia Ecclesiastica (HE)
v.8.3 106
v.16.3 107

Ignatius
Philadelphians
9.2 54–55
Smyrnaeans
5.1 79
7.2 55, 79

Irenaeus
Adversus Haereses
ii.22.3 106
ii.27.2 105
iii Preface 15, 65, 108
iii.1.1–2 66
iii.1.1 65, 106
iii.2.1 66
iii.9.1 106
iii.9.12 106
iii.10.1 105
iii.10.3 67–70
 Index of references 229
iii.10.6 105
iii.11.1 105
iii.11.7 87
iii.11.8 64–66, 91, 106
iii.11.9 66–67, 108
iii.11.12 106
iii.14.3–4 80
iv Preface 108
iv.1.1 108
iv.6.1 108
iv.3 67
iv.7 67
iv.20.6 106
iv.41.2 106

Justin Martyr
I Apology
4.7 89
6.1 95
6.2 95
14 95
14.4 95, 96
14.5 96
15–17 96
15.1–4 96
15.10–17 96
17.1–2 96
26 138
30 135, 145
54.12 104
56 138
61–3 97
61.4–5 102
61.4 76
65.5 95
66 75
66.1 95
66.2 95
66.3 95, 99, 100
67 204
67.3 99, 100
67.3–4 54, 75

II Apology
3.3 103

Dialogue
7.3 138
8.1 96
8.2 77–78, 97
10.2 76, 97, 100, 103
10.3 118
11.2 118
230 Index of references
Dialogue (cont.)
11.4 118
12.2 118
14.3 118
17 154
17.3 98
18.1 97, 98, 103
35.3 96, 154
38.1 153
43.1 118
45.2–3 153, 155
47.5 103
51.2 154
56.12 98, 105
56.17 98, 105
61.4–8 97
62.3 97, 154
63.1 97
63.3–5 97
69 137
69.7 130, 131, 132, 140, 146
78.1 98
78.5 103
80.3–4 153, 154
81.4 101
88.3 97, 100, 103
98–107 76
100–107 54, 97, 99
100.1 76
100.2 100
103.8 76, 78, 100, 103, 105
106.3 76, 101
106.4 101
108 135, 140, 154, 155
113–14 97
113.6–7 97
114.4 97
119.6 98
134.1 99

Kerygma Petri 117

Muratorian Fragment 67–68, 71, 74, 78, 79, 80, 89

Origen
Contra Celsum
i.28 145, 150
i.42 152
i.51 153
ii.68 138
ii.16 150
ii.27 67–70, 152
ii.49 138
ii.55–8 151
ii.55 150
 Index of references 231
ii.60 152
ii.63 152
ii.67 152
ii.72 152
ii.77 150, 152
iv.56 150
v.14–26 150
v.14 150
v.52 150
v.55–6 152
v.56 70
vi.29 150
vii.68 150

Shepherd of Hermas
5.6.3 (= ch. 59) 117
8.3.2–3 (= ch. 69) 117

Graeco-Roman writings

Aeschylus
Eumenides 647–8 156

Cicero
Epistulae ad Familiares 9.26.1 185

Dio Cassius 59.28 24

Lucian of Samosata
Peregrinus 13 116

Martial
Epigrams
i.2 84–91, 179
xiv.4–7 173
xiv.184–92 179

Philostratus
Life 5.12 136

Quintilian
Institutiones Oratoriae
x.3.31–2 177

Suetonius
Caligula 22 24

Inscriptions and papyri

Inscriptions
I. Olympia 53 32
I. Priene (Calendar inscription) 30–32
Pisidian Antioch (temple) 38
232 Index of references
Papyri
∏1 197
∏4 97
∏45 72, 74, 96, 116, 197, 201, 202
∏52 83, 106, 158, 206
∏53 198
∏64 97
∏64 + ∏67 + ∏4 74, 78, 116, 117, 198, 205, 206
∏66 78, 100, 201, 206
∏67 97
∏69 203
∏75 72, 74, 78, 97, 100, 116, 201, 203
∏84 202–203
∏88 76
∏90 201, 206
∏95 201, 202
∏103 199
∏104 199, 201
∏106 201
∏107 201
∏108 201
∏109 201
∏111 203
P. Oxy. 2 197
P. Oxy. 30 158, 159, 179, 204
P. Oxy. 405 204
P. Oxy. 2192 195
P. Oxy. 2383 203
P. Oxy. 2683 199, 206
P. Oxy. 3523 201, 206
P. Oxy. 4403 199
P. Oxy. 4404 199, 201
P. Oxy. 4405 199
P. Oxy. 4447 201
P. Oxy. 4442 205
P. Oxy. 4443 205
P. Oxy. 4495 203
P. Petaus 30 180, 181, 184, 189–190
Papyri Graecae Magicae (PGM) 5.110 137
Papyrus Rylands Greek (PRG) 460 153, 183
 General index

Abramowski, Luise 75, 101

Acts of Thomas 132–133
Adamnan, abbot of Iona 174
Albl, Martin 184
Aldhelm, bishop of Sherborne 174
Alexander, Loveday 196
Alexandria 167
Alföldy, Geza 27
Alogi 70
Anabathmoi Jacobou 71
Ancyra 36
Andrew 70
Antioch on the Orontes 23
Antioch, Pisidian 12, 31, 32, 36, 37–39, 46
Aphrodisias 66
apostolic fathers 191
Aquinas 111
Aramaic 18, 21
Aristotle 107, 195
Augustus 24, 26, 27–30, 38–39, 40, 42, 47
Res Gestae 30, 36, 39, 42

Baarda, T. 89
Bar Kokhba 85, 140, 175
Barclay, John 44
Barnabas, Epistle of 118
Barr, James 2
Barth, K. 112
Bauckham, R. J. 193
Beatitudes 14–15
benefactor 25, 32, 46
Bengel, J. A. 121
Betz, H. D. 41, 110
‘big bang’ theories 81, 85, 167–169, 181
biographies 192
Blanchard, Y.-M. 107
bookhands 194, 205
Brooke, George 183
Burton, E. C. 119

233
234 General index
Calendar (Priene) inscription 30–32
Caligula (Gaius) 23, 24–25, 28, 46
Campenhausen, H. von 63, 71, 81, 107
Celsus 70, 135, 138, 145
Chambers, J. K. 50
Charlesworth, J. H. 128
Chartres 119
Chester Beatty papyri 71, 82, 165–167, 196, 202–203
Chrysostom, J. 156
Cicero 185, 191
Claudius 23, 46
Clement of Alexandria 53, 117
II Clement 79, 92
Cockle, W. E. H. 202, 203
codex 5–6, 94
double columns 73
four-gospel codices 71–75
Codex Bezae 156
Codex Sinaiticus 74, 195, 201
Codex Vaticanus 73, 195
colonies, Roman 36, 37–38
Cosgrove, C. H. 93
Crossan, J. D. 88
Cullmann, O. 64
Cyprian 90

Davies, W. D. 120
De Bellis Macedonicis 179
Deissmann, A. 30
demonic possession 129, 141
Derbe 36
Dibelius, Martin 59
Didache 55, 79, 92
documentary hands 194–195, 205
Dodd, C. H. 120
Domitian 28
double columns, see codex
Durkheim, E. 50

Ehrman, B. 196
Eisler, Robert 133
Epiphanius 138
Epp, E. J. 169, 171, 172
Erasmus 111
Eusebius 107
exorcism 140, 142, 143, 144–146

Falcetta, A. 183
Feldman, L. H. 133–141
Fisher, John 12
Fourth Gospel 69, 70, 76, 78, 79, 83, 87, 89, 90, 101–102, 105,
168, 200–202, 204
Frankemölle, H. 2
Friedrich, G. 1, 29
Furnish, V. 121
 General index 235
Gaius (anti-Montanist) 70
Gaius (Caligula) 23, 24–25, 28, 46
Galatians 39–45, 51
agitators 43–45
Christology 40
Galatian churches 22
Galen of Pergamum 149
Gamble, Harry 168, 174, 196, 205
genre 192
Gerhardsson, B. 187
Gospel, fourfold 4, 105
Gospel of Peter 87, 88, 109
Gospel of Thomas 3–4, 87, 109
Gospel of Truth 88
gospels as biographies 192
Guelich, Robert 145
Gutenberg, Johannes 165–191

Haelst, Joseph van 171, 174, 179

Hahneman, G. 68
Harnack, A. 63, 78, 101, 130
Harris, J. Rendel 182
Harris, W. V. 170–171, 188
Harrison, J. R. 47, 51
Hatch, E. 182
Hays, Richard 110
Hellenistic ruler cults 27
Hengel, Martin 54, 56, 78–79
Herculaneum 173
Hill, Charles 98
Hippolytus 71, 90
Hooker, M. D. ix
Horsley, G. H. R. 33
Hyldahl, N. 104

Iconium 12, 36, 37, 46

Ignatius 79, 92
imperial cult 2, 5, 11, 24–35, 38–45, 58, 59
inscriptions 20
Irenaeus 4, 53, 58, 63–68, 69, 70, 71, 74, 75, 78, 80, 81, 86,
87, 88, 90, 91, 166, 206
Isaiah 97

James 115
Jerusalem 23
Jesus
Cynic 86
false prophet / deceiver 17, 127–147
magician / sorceror 5, 17, 127–147
messianic prophet 17
prophet 146
self-understanding 13, 17, 22
wisdom teacher 86
Jesus Seminar 88
I John 116
236 General index
John’s Gospel 69, 70, 76, 78, 79, 83, 87, 89, 90, 101–102,
105, 168, 200–202, 204
John the Baptist 146
Josephus 29–30, 33, 128, 133, 136, 146
Testimonium Flavianum 133–134, 137
Justin Martyr 4, 12, 71, 85, 92, 93–109, 122, 129–130, 131, 137–138,
140, 145, 153–155, 191
memoirs of the apostles 95, 97, 99–105, 204–205

Kenyon, F. G. 71–72, 82, 109, 165–167, 191, 198

Kerygma Petri 117, 122
Kingsbury, J. D. 57
Koester, H. 54, 77, 78, 103

Labov, William 50
Lapide, Pinchas 159
law of Christ 4–5, 110–123
law of Moses 112–115, 119, 122
lectionaries 90
Levick, Barbara 37
Liebermann, S. 122, 167, 186–187
Lightfoot, J. B. 94
Lim, Timothy 184
literacy levels 31
Longenecker, R. N. 120
love commandment 116–117
Luchner, K. 205
Lucian of Samosata 136, 149–161
Luke as evangelist 14, 60
Luke’s Gospel 69, 70, 80–81, 87, 141, 190
papyri 202–203
Luke–Acts 80–81, 166
Luther, Martin 88, 111, 121
Luz, Ulrich 188
LXX (Septuagint) 21–22, 35, 52
Lystra 12, 36, 37, 45, 46

Marcion 12, 53, 63, 81, 87

Marcovich, Miroslav 93, 101
Marcus Aurelius 94
Mark’s Gospel 12, 76, 87, 101, 106, 108, 143, 167, 190
papyri 202
use of to eÉagg”lion 18–20, 55–58, 60
Martial 84, 170, 173, 179, 180, 190, 206
Marxsen, Willi 18
Matthew as evangelist 4, 12, 56–58, 60
Matthew’s Gospel 71, 87, 98, 115–116, 204
papyri 197–200
Meeks, Wayne 49
Meggitt, Justin 28, 33, 35
Meier, J. P. 133, 134
memoirs of the apostles, see Justin Martyr
Millard, A. R. 188
Milroy, Lesley 50
Mitchell, Margaret 49
 General index 237
Mitchell, Stephen 28, 36, 37, 38, 42
Montanus 138
Moule, C. F. D. ix, x
Muddiman, John 186
Muratorian Fragment 63–64, 67, 68–71, 74, 75, 78, 79, 80, 89

Neusner, Jacob 187

Nicolaus of Damascus 26
nomina sacra 64, 83, 190, 205, 206
notebooks 6, 165, 170, 175, 176–178, 181–191

Oakes, Peter 48–49

O’Callaghan, J. 202–203
oral traditions 85, 108
Origen 70, 71, 138, 149–153
Otto, J. C. Th. Von 101
Oxyrhynchus papyri 193, 204

Paphos 37
Papias 70, 79, 108
papyri 5, 6, 21, 32
Parisinus codex (Justin Martyr) 101
parousias, two 71
Parsons, Peter 199
Perge 37
Pessinus 36
Peter 101
Petersen, W. L. 77, 89
Petronius 24
Philippi 12, 47, 48–49
Philippians 48–49, 51
Philo 24, 26, 28–30, 46, 136
Philostratus 136
Pisidian Antioch 12, 31, 32, 36, 37–39, 46
Pompeii 173
Pope John Paul II 10
Priene (Calendar) inscription 30–32
Pryor, J. W. 102

Q traditions 2–4, 15, 18, 138, 144, 188, 193

Q as ‘gospel’ 3
Quintilian 170, 177, 178, 191

rabbinic traditions 21, 130–132

Ramsay, W. M. 38
Reimarus, H. S. 127
Res Gestae (Augustus) 30, 36, 39, 42
resurrection 5, 148
Riches, J. 128
Roberts, Colin 72, 74, 82, 170–171, 179, 183, 187, 194, 195, 198, 199,
200, 201, 202–203, 205, 206
roll format 165–166, 172–173, 181, 189
Roman colonies 36, 37–38
Rome 96, 105, 109, 167, 168
ruler cults (Hellenistic) 27
238 General index
salvation 45, 47
Sanders, E. P. 128
Sanders, J. N. 136
Sato, Migaku 188
Saviour 25, 32, 47, 48
Schnackenburg, R. 90, 91
Schniewind, J. 1
Scripture 105–106
Second Vatican Council 59
semantic field 12, 51, 53, 61
Sepphoris 88
Serapion 88
Shepherd of Hermas 68, 117, 118, 122
shorthand (language) 48, 49–50, 61, 114
Sirat, Colette 187
Skeat, T. C. 177, 179, 187
social dialect 52, 61
sociolect 11, 12
sociolinguistics 2, 11, 50, 52
Socrates 104
Spirit 70, 71, 89, 97, 145
Strange, W. A. 80
Strauss, D. F. 127
Strecker, G. 2
‘street’ language 1, 2, 10, 52
Stuhlmacher, P. 19
Sundberg, A. 68
Sweet, J. P. M. 139
synoptic gospels 89, 90

tablets (leaf and stylus) 174–176

tablets (wax and wooden) 173–176
Tatian 75, 77, 78, 87, 89, 103, 109
Tavium 36
Tertullian 53, 71
Testament of Levi 135
Theodore of Mopsuestia 89
I and II Thessalonians 47–48, 51
Thessalonica 12, 22, 47–48
Thiede, C. P. 73, 198, 203
Thomas, J. D. 199
Toledoth Jeshu 154
Trobisch, D. 83
Trypho (Justin Martyr) 76, 77, 153–155
Turner, Eric 194, 196, 203
Tyndale, William 9

Unnik, W. C. van 107

Valentinians 66, 70, 81, 88

Vermes, G. 128
Vespasian 29
Victorinus of Pettau 90
Vindolanda 173, 174–176, 178
 General index 239
Weber, Max 50
Winter, B. W. 43–45
Witulski, T. 41–43
word (logos) 47–48, 49, 51, 60–61
worship 99

Xenophon, Memorabilia 104

Yadin, Y. 175

Zahn, Th. 63, 78, 101